YOLOv3.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "YOLOv3.ipynb",
      "provenance": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/robinjacobroy/Odonates/blob/master/YOLOv3.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CHSXDMODWfoN",
        "colab_type": "code",
        "outputId": "0e4ae826-2fd4-47bc-cfd8-9cbd0a13093f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vwOpjGP7-Nhr",
        "colab_type": "code",
        "outputId": "2bfef2dc-4a37-4ada-a286-46e2398f1b21",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "%cd \"/content/gdrive/My Drive/Odonata_data/YOLO/darknet-master/\"\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/gdrive/My Drive/Odonata_data/YOLO/darknet-master\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qzDdPWsG7HN2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!cp ../yolov3-odonata.cfg ."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rs38oIB-0HjK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!find . -type f -mtime -.208 -exec rm -- {} +"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X2Fum8u0CR0J",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!mv train_field2/* darknet-master/data/odonata/"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QQFI1Hke-XMD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!for file in *.txt ; do mv $file ${file//(1)/_1} ; done"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b2ax1kLIZw1W",
        "colab_type": "code",
        "outputId": "1b8c07cc-01b8-4e67-835c-43b885dd46cd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "#!ls data/odonata/ | wc -l"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2722\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v_ecbVgs1IuK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "cp ../Makefile ."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uppl2kNnZWnt",
        "colab_type": "code",
        "outputId": "aaf39207-309e-4353-931a-1fd553f47a38",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#!cmake . && make\n",
        "!make"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "chmod +x *.sh\n",
            "g++ -std=c++11 -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/image_opencv.cpp -o obj/image_opencv.o\n",
            "g++ -std=c++11 -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/http_stream.cpp -o obj/http_stream.o\n",
            "\u001b[01m\u001b[K./src/http_stream.cpp:\u001b[m\u001b[K In member function ‘\u001b[01m\u001b[Kbool JSON_sender::write(const char*)\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/http_stream.cpp:228:21:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kn\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "                 int \u001b[01;35m\u001b[Kn\u001b[m\u001b[K = _write(client, outputbuf, outlen);\n",
            "                     \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/gemm.c -o obj/gemm.o\n",
            "\u001b[01m\u001b[K./src/gemm.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kconvolution_2d\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/gemm.c:2037:15:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kout_w\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     const int \u001b[01;35m\u001b[Kout_w\u001b[m\u001b[K = (w + 2 * pad - ksize) / stride + 1;    // output_width=input_width for stride=1 and pad=1\n",
            "               \u001b[01;35m\u001b[K^~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/gemm.c:2036:15:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kout_h\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     const int \u001b[01;35m\u001b[Kout_h\u001b[m\u001b[K = (h + 2 * pad - ksize) / stride + 1;    // output_height=input_height for stride=1 and pad=1\n",
            "               \u001b[01;35m\u001b[K^~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/utils.c -o obj/utils.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/dark_cuda.c -o obj/dark_cuda.o\n",
            "\u001b[01m\u001b[K./src/dark_cuda.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kget_cuda_stream\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/dark_cuda.c:120:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kbuffer\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "             char \u001b[01;35m\u001b[Kbuffer\u001b[m\u001b[K[256];\n",
            "                  \u001b[01;35m\u001b[K^~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/dark_cuda.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kget_cuda_memcpy_stream\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/dark_cuda.c:141:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kbuffer\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "             char \u001b[01;35m\u001b[Kbuffer\u001b[m\u001b[K[256];\n",
            "                  \u001b[01;35m\u001b[K^~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/dark_cuda.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kcudnn_handle\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/dark_cuda.c:161:23:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kstatus\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "         cudnnStatus_t \u001b[01;35m\u001b[Kstatus\u001b[m\u001b[K = cudnnSetStream(handle[i], get_cuda_stream());\n",
            "                       \u001b[01;35m\u001b[K^~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/convolutional_layer.c -o obj/convolutional_layer.o\n",
            "\u001b[01m\u001b[K./src/convolutional_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kforward_convolutional_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/convolutional_layer.c:1019:32:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kt_intput_size\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "                         size_t \u001b[01;35m\u001b[Kt_intput_size\u001b[m\u001b[K = binary_transpose_align_input(k, n, state.workspace, &l.t_bit_input, ldb_align, l.bit_align);\n",
            "                                \u001b[01;35m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/convolutional_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kassisted_excitation_forward\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/convolutional_layer.c:1101:17:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kepoch\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     const float \u001b[01;35m\u001b[Kepoch\u001b[m\u001b[K = (float)(*state.net.seen) / state.net.train_images_num;\n",
            "                 \u001b[01;35m\u001b[K^~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/list.c -o obj/list.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/image.c -o obj/image.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/activations.c -o obj/activations.o\n",
            "\u001b[01m\u001b[K./src/activations.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kactivate\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/activations.c:68:5:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kenumeration value ‘\u001b[01m\u001b[KSWISH\u001b[m\u001b[K’ not handled in switch [\u001b[01;35m\u001b[K-Wswitch\u001b[m\u001b[K]\n",
            "     \u001b[01;35m\u001b[Kswitch\u001b[m\u001b[K(a){\n",
            "     \u001b[01;35m\u001b[K^~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/activations.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kgradient\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/activations.c:138:5:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kenumeration value ‘\u001b[01m\u001b[KSWISH\u001b[m\u001b[K’ not handled in switch [\u001b[01;35m\u001b[K-Wswitch\u001b[m\u001b[K]\n",
            "     \u001b[01;35m\u001b[Kswitch\u001b[m\u001b[K(a){\n",
            "     \u001b[01;35m\u001b[K^~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/im2col.c -o obj/im2col.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/col2im.c -o obj/col2im.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/blas.c -o obj/blas.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/crop_layer.c -o obj/crop_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/dropout_layer.c -o obj/dropout_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/maxpool_layer.c -o obj/maxpool_layer.o\n",
            "\u001b[01m\u001b[K./src/maxpool_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kcudnn_maxpool_setup\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/maxpool_layer.c:26:19:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kvariable ‘\u001b[01m\u001b[Kmaxpool_status\u001b[m\u001b[K’ set but not used [\u001b[01;35m\u001b[K-Wunused-but-set-variable\u001b[m\u001b[K]\n",
            "     cudnnStatus_t \u001b[01;35m\u001b[Kmaxpool_status\u001b[m\u001b[K;\n",
            "                   \u001b[01;35m\u001b[K^~~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/softmax_layer.c -o obj/softmax_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/data.c -o obj/data.o\n",
            "\u001b[01m\u001b[K./src/data.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kload_data_detection\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/data.c:1046:19:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kscale\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "             float \u001b[01;35m\u001b[Kscale\u001b[m\u001b[K = rand_precalc_random(.25, 2, r_scale); // unused currently\n",
            "                   \u001b[01;35m\u001b[K^~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/data.c:1046:19:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K‘\u001b[01m\u001b[Kr_scale\u001b[m\u001b[K’ may be used uninitialized in this function [\u001b[01;35m\u001b[K-Wmaybe-uninitialized\u001b[m\u001b[K]\n",
            "             float \u001b[01;35m\u001b[Kscale\u001b[m\u001b[K = rand_precalc_random(.25, 2, r_scale); // unused currently\n",
            "                   \u001b[01;35m\u001b[K^~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/matrix.c -o obj/matrix.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/network.c -o obj/network.o\n",
            "\u001b[01m\u001b[K./src/network.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kresize_network\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/network.c:562:42:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kpassing argument 1 of ‘\u001b[01m\u001b[KcudaHostAlloc\u001b[m\u001b[K’ from incompatible pointer type [\u001b[01;35m\u001b[K-Wincompatible-pointer-types\u001b[m\u001b[K]\n",
            "         if (cudaSuccess == cudaHostAlloc(\u001b[01;35m\u001b[K&\u001b[m\u001b[Knet->input_pinned_cpu, size * sizeof(float), cudaHostRegisterMapped))\n",
            "                                          \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "In file included from \u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime.h:96:0\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[Kinclude/darknet.h:39\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/network.c:1\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime_api.h:4412:39:\u001b[m\u001b[K \u001b[01;36m\u001b[Knote: \u001b[m\u001b[Kexpected ‘\u001b[01m\u001b[Kvoid **\u001b[m\u001b[K’ but argument is of type ‘\u001b[01m\u001b[Kfloat **\u001b[m\u001b[K’\n",
            " extern __host__ cudaError_t CUDARTAPI \u001b[01;36m\u001b[KcudaHostAlloc\u001b[m\u001b[K(void **pHost, size_t size, unsigned int flags);\n",
            "                                       \u001b[01;36m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/connected_layer.c -o obj/connected_layer.o\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kforward_connected_layer_gpu\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:335:11:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kone\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     float \u001b[01;35m\u001b[Kone\u001b[m\u001b[K = 1;    // alpha[0], beta[0]\n",
            "           \u001b[01;35m\u001b[K^~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:333:13:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kc\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     float * \u001b[01;35m\u001b[Kc\u001b[m\u001b[K = l.output_gpu;\n",
            "             \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:332:13:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kb\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     float * \u001b[01;35m\u001b[Kb\u001b[m\u001b[K = l.weights_gpu;\n",
            "             \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:331:13:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Ka\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     float * \u001b[01;35m\u001b[Ka\u001b[m\u001b[K = state.input;\n",
            "             \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:330:9:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kn\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     int \u001b[01;35m\u001b[Kn\u001b[m\u001b[K = l.outputs;\n",
            "         \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:329:9:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kk\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     int \u001b[01;35m\u001b[Kk\u001b[m\u001b[K = l.inputs;\n",
            "         \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/connected_layer.c:328:9:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Km\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     int \u001b[01;35m\u001b[Km\u001b[m\u001b[K = l.batch;\n",
            "         \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/cost_layer.c -o obj/cost_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/parser.c -o obj/parser.o\n",
            "\u001b[01m\u001b[K./src/parser.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kparse_network_cfg_custom\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/parser.c:1007:42:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kpassing argument 1 of ‘\u001b[01m\u001b[KcudaHostAlloc\u001b[m\u001b[K’ from incompatible pointer type [\u001b[01;35m\u001b[K-Wincompatible-pointer-types\u001b[m\u001b[K]\n",
            "         if (cudaSuccess == cudaHostAlloc(\u001b[01;35m\u001b[K&\u001b[m\u001b[Knet.input_pinned_cpu, size * sizeof(float), cudaHostRegisterMapped)) net.input_pinned_cpu_flag = 1;\n",
            "                                          \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "In file included from \u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime.h:96:0\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[Kinclude/darknet.h:39\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/activations.h:3\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/activation_layer.h:4\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/parser.c:6\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime_api.h:4412:39:\u001b[m\u001b[K \u001b[01;36m\u001b[Knote: \u001b[m\u001b[Kexpected ‘\u001b[01m\u001b[Kvoid **\u001b[m\u001b[K’ but argument is of type ‘\u001b[01m\u001b[Kfloat **\u001b[m\u001b[K’\n",
            " extern __host__ cudaError_t CUDARTAPI \u001b[01;36m\u001b[KcudaHostAlloc\u001b[m\u001b[K(void **pHost, size_t size, unsigned int flags);\n",
            "                                       \u001b[01;36m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/option_list.c -o obj/option_list.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/darknet.c -o obj/darknet.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/detection_layer.c -o obj/detection_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/captcha.c -o obj/captcha.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/route_layer.c -o obj/route_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/writing.c -o obj/writing.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/box.c -o obj/box.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/nightmare.c -o obj/nightmare.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/normalization_layer.c -o obj/normalization_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/avgpool_layer.c -o obj/avgpool_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/coco.c -o obj/coco.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/dice.c -o obj/dice.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/yolo.c -o obj/yolo.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/detector.c -o obj/detector.o\n",
            "\u001b[01m\u001b[K./src/detector.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Ktrain_detector\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/detector.c:52:16:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Knames\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "         char **\u001b[01;35m\u001b[Knames\u001b[m\u001b[K = get_labels_custom(name_list, &names_size);\n",
            "                \u001b[01;35m\u001b[K^~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/detector.c:258:13:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kvariable ‘\u001b[01m\u001b[Kdraw_precision\u001b[m\u001b[K’ set but not used [\u001b[01;35m\u001b[K-Wunused-but-set-variable\u001b[m\u001b[K]\n",
            "         int \u001b[01;35m\u001b[Kdraw_precision\u001b[m\u001b[K = 0;\n",
            "             \u001b[01;35m\u001b[K^~~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/detector.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kvalidate_detector_map\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/detector.c:1046:15:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kclass_recall\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "         float \u001b[01;35m\u001b[Kclass_recall\u001b[m\u001b[K = (float)tp_for_thresh_per_class[i] / ((float)tp_for_thresh_per_class[i] + (float)(truth_classes_count[i] - tp_for_thresh_per_class[i]));\n",
            "               \u001b[01;35m\u001b[K^~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/detector.c:1045:15:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kclass_precision\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "         float \u001b[01;35m\u001b[Kclass_precision\u001b[m\u001b[K = (float)tp_for_thresh_per_class[i] / ((float)tp_for_thresh_per_class[i] + (float)fp_for_thresh_per_class[i]);\n",
            "               \u001b[01;35m\u001b[K^~~~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/layer.c -o obj/layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/compare.c -o obj/compare.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/classifier.c -o obj/classifier.o\n",
            "\u001b[01m\u001b[K./src/classifier.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Ktrain_classifier\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/classifier.c:157:13:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kvariable ‘\u001b[01m\u001b[Kdraw_precision\u001b[m\u001b[K’ set but not used [\u001b[01;35m\u001b[K-Wunused-but-set-variable\u001b[m\u001b[K]\n",
            "         int \u001b[01;35m\u001b[Kdraw_precision\u001b[m\u001b[K = 0;\n",
            "             \u001b[01;35m\u001b[K^~~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/local_layer.c -o obj/local_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/swag.c -o obj/swag.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/shortcut_layer.c -o obj/shortcut_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/activation_layer.c -o obj/activation_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/rnn_layer.c -o obj/rnn_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/gru_layer.c -o obj/gru_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/rnn.c -o obj/rnn.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/rnn_vid.c -o obj/rnn_vid.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/crnn_layer.c -o obj/crnn_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/demo.c -o obj/demo.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/tag.c -o obj/tag.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/cifar.c -o obj/cifar.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/go.c -o obj/go.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/batchnorm_layer.c -o obj/batchnorm_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/art.c -o obj/art.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/region_layer.c -o obj/region_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/reorg_layer.c -o obj/reorg_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/reorg_old_layer.c -o obj/reorg_old_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/super.c -o obj/super.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/voxel.c -o obj/voxel.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/tree.c -o obj/tree.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/yolo_layer.c -o obj/yolo_layer.o\n",
            "\u001b[01m\u001b[K./src/yolo_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kmake_yolo_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/yolo_layer.c:58:38:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kpassing argument 1 of ‘\u001b[01m\u001b[KcudaHostAlloc\u001b[m\u001b[K’ from incompatible pointer type [\u001b[01;35m\u001b[K-Wincompatible-pointer-types\u001b[m\u001b[K]\n",
            "     if (cudaSuccess == cudaHostAlloc(\u001b[01;35m\u001b[K&\u001b[m\u001b[Kl.output, batch*l.outputs*sizeof(float), cudaHostRegisterMapped)) l.output_pinned = 1;\n",
            "                                      \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "In file included from \u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime.h:96:0\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[Kinclude/darknet.h:39\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/activations.h:3\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/layer.h:4\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.h:5\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.c:1\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime_api.h:4412:39:\u001b[m\u001b[K \u001b[01;36m\u001b[Knote: \u001b[m\u001b[Kexpected ‘\u001b[01m\u001b[Kvoid **\u001b[m\u001b[K’ but argument is of type ‘\u001b[01m\u001b[Kfloat **\u001b[m\u001b[K’\n",
            " extern __host__ cudaError_t CUDARTAPI \u001b[01;36m\u001b[KcudaHostAlloc\u001b[m\u001b[K(void **pHost, size_t size, unsigned int flags);\n",
            "                                       \u001b[01;36m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/yolo_layer.c:65:38:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kpassing argument 1 of ‘\u001b[01m\u001b[KcudaHostAlloc\u001b[m\u001b[K’ from incompatible pointer type [\u001b[01;35m\u001b[K-Wincompatible-pointer-types\u001b[m\u001b[K]\n",
            "     if (cudaSuccess == cudaHostAlloc(\u001b[01;35m\u001b[K&\u001b[m\u001b[Kl.delta, batch*l.outputs*sizeof(float), cudaHostRegisterMapped)) l.delta_pinned = 1;\n",
            "                                      \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "In file included from \u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime.h:96:0\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[Kinclude/darknet.h:39\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/activations.h:3\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/layer.h:4\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.h:5\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.c:1\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime_api.h:4412:39:\u001b[m\u001b[K \u001b[01;36m\u001b[Knote: \u001b[m\u001b[Kexpected ‘\u001b[01m\u001b[Kvoid **\u001b[m\u001b[K’ but argument is of type ‘\u001b[01m\u001b[Kfloat **\u001b[m\u001b[K’\n",
            " extern __host__ cudaError_t CUDARTAPI \u001b[01;36m\u001b[KcudaHostAlloc\u001b[m\u001b[K(void **pHost, size_t size, unsigned int flags);\n",
            "                                       \u001b[01;36m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/yolo_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kresize_yolo_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/yolo_layer.c:92:42:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kpassing argument 1 of ‘\u001b[01m\u001b[KcudaHostAlloc\u001b[m\u001b[K’ from incompatible pointer type [\u001b[01;35m\u001b[K-Wincompatible-pointer-types\u001b[m\u001b[K]\n",
            "         if (cudaSuccess != cudaHostAlloc(\u001b[01;35m\u001b[K&\u001b[m\u001b[Kl->output, l->batch*l->outputs * sizeof(float), cudaHostRegisterMapped)) {\n",
            "                                          \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "In file included from \u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime.h:96:0\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[Kinclude/darknet.h:39\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/activations.h:3\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/layer.h:4\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.h:5\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.c:1\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime_api.h:4412:39:\u001b[m\u001b[K \u001b[01;36m\u001b[Knote: \u001b[m\u001b[Kexpected ‘\u001b[01m\u001b[Kvoid **\u001b[m\u001b[K’ but argument is of type ‘\u001b[01m\u001b[Kfloat **\u001b[m\u001b[K’\n",
            " extern __host__ cudaError_t CUDARTAPI \u001b[01;36m\u001b[KcudaHostAlloc\u001b[m\u001b[K(void **pHost, size_t size, unsigned int flags);\n",
            "                                       \u001b[01;36m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/yolo_layer.c:101:42:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kpassing argument 1 of ‘\u001b[01m\u001b[KcudaHostAlloc\u001b[m\u001b[K’ from incompatible pointer type [\u001b[01;35m\u001b[K-Wincompatible-pointer-types\u001b[m\u001b[K]\n",
            "         if (cudaSuccess != cudaHostAlloc(\u001b[01;35m\u001b[K&\u001b[m\u001b[Kl->delta, l->batch*l->outputs * sizeof(float), cudaHostRegisterMapped)) {\n",
            "                                          \u001b[01;35m\u001b[K^\u001b[m\u001b[K\n",
            "In file included from \u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime.h:96:0\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[Kinclude/darknet.h:39\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/activations.h:3\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/layer.h:4\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.h:5\u001b[m\u001b[K,\n",
            "                 from \u001b[01m\u001b[K./src/yolo_layer.c:1\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K/usr/local/cuda/include/cuda_runtime_api.h:4412:39:\u001b[m\u001b[K \u001b[01;36m\u001b[Knote: \u001b[m\u001b[Kexpected ‘\u001b[01m\u001b[Kvoid **\u001b[m\u001b[K’ but argument is of type ‘\u001b[01m\u001b[Kfloat **\u001b[m\u001b[K’\n",
            " extern __host__ cudaError_t CUDARTAPI \u001b[01;36m\u001b[KcudaHostAlloc\u001b[m\u001b[K(void **pHost, size_t size, unsigned int flags);\n",
            "                                       \u001b[01;36m\u001b[K^~~~~~~~~~~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/upsample_layer.c -o obj/upsample_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/lstm_layer.c -o obj/lstm_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/conv_lstm_layer.c -o obj/conv_lstm_layer.o\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/scale_channels_layer.c -o obj/scale_channels_layer.o\n",
            "In file included from \u001b[01m\u001b[K./src/scale_channels_layer.c:5:0\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K./src/scale_channels_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kmake_scale_channels_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/scale_channels_layer.c:16:14:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Ksuggest parentheses around comparison in operand of ‘\u001b[01m\u001b[K&\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wparentheses\u001b[m\u001b[K]\n",
            "     assert(\u001b[01;35m\u001b[Kw == 1\u001b[m\u001b[K & h == 1);\n",
            "            \u001b[01;35m\u001b[K~~^~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/scale_channels_layer.c:16:14:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Ksuggest parentheses around comparison in operand of ‘\u001b[01m\u001b[K&\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wparentheses\u001b[m\u001b[K]\n",
            "     assert(\u001b[01;35m\u001b[Kw == 1\u001b[m\u001b[K & h == 1);\n",
            "            \u001b[01;35m\u001b[K~~^~~~\u001b[m\u001b[K\n",
            "gcc -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include -c ./src/sam_layer.c -o obj/sam_layer.o\n",
            "In file included from \u001b[01m\u001b[K./src/sam_layer.c:5:0\u001b[m\u001b[K:\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kmake_sam_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:21:16:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Ksuggest parentheses around comparison in operand of ‘\u001b[01m\u001b[K&\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wparentheses\u001b[m\u001b[K]\n",
            "     assert(\u001b[01;35m\u001b[Kl.w == l.o\u001b[m\u001b[Kut_w & l.h == l.out_h);\n",
            "            \u001b[01;35m\u001b[K~~~~^~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:21:16:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Ksuggest parentheses around comparison in operand of ‘\u001b[01m\u001b[K&\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wparentheses\u001b[m\u001b[K]\n",
            "     assert(\u001b[01;35m\u001b[Kl.w == l.o\u001b[m\u001b[Kut_w & l.h == l.out_h);\n",
            "            \u001b[01;35m\u001b[K~~~~^~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kforward_sam_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:63:9:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kchannel_size\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     int \u001b[01;35m\u001b[Kchannel_size\u001b[m\u001b[K = 1;\n",
            "         \u001b[01;35m\u001b[K^~~~~~~~~~~~\u001b[m\u001b[K\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:\u001b[m\u001b[K In function ‘\u001b[01m\u001b[Kbackward_sam_layer\u001b[m\u001b[K’:\n",
            "\u001b[01m\u001b[K./src/sam_layer.c:82:9:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[Kunused variable ‘\u001b[01m\u001b[Kchannel_size\u001b[m\u001b[K’ [\u001b[01;35m\u001b[K-Wunused-variable\u001b[m\u001b[K]\n",
            "     int \u001b[01;35m\u001b[Kchannel_size\u001b[m\u001b[K = 1;\n",
            "         \u001b[01;35m\u001b[K^~~~~~~~~~~~\u001b[m\u001b[K\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/convolutional_kernels.cu -o obj/convolutional_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/activation_kernels.cu -o obj/activation_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/im2col_kernels.cu -o obj/im2col_kernels.o\n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:125:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "                 //*data_col_ptr = (h >= 0 && w >= 0 && h < height && w < width) ?\n",
            "                   \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:1178:6:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "     //*((uint64_t *)(A_s + (local_i*lda + k) / 8)) = *((uint64_t *)(A + (i_cur*lda + k) / 8));    // weights\n",
            "       \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:125:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "                 //*data_col_ptr = (h >= 0 && w >= 0 && h < height && w < width) ?\n",
            "                   \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:1178:6:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "     //*((uint64_t *)(A_s + (local_i*lda + k) / 8)) = *((uint64_t *)(A + (i_cur*lda + k) / 8));    // weights\n",
            "       \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:125:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "                 //*data_col_ptr = (h >= 0 && w >= 0 && h < height && w < width) ?\n",
            "                   \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:1178:6:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "     //*((uint64_t *)(A_s + (local_i*lda + k) / 8)) = *((uint64_t *)(A + (i_cur*lda + k) / 8));    // weights\n",
            "       \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:125:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "                 //*data_col_ptr = (h >= 0 && w >= 0 && h < height && w < width) ?\n",
            "                   \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:1178:6:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "     //*((uint64_t *)(A_s + (local_i*lda + k) / 8)) = *((uint64_t *)(A + (i_cur*lda + k) / 8));    // weights\n",
            "       \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:125:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "                 //*data_col_ptr = (h >= 0 && w >= 0 && h < height && w < width) ?\n",
            "                   \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:1178:6:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "     //*((uint64_t *)(A_s + (local_i*lda + k) / 8)) = *((uint64_t *)(A + (i_cur*lda + k) / 8));    // weights\n",
            "       \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:125:18:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "                 //*data_col_ptr = (h >= 0 && w >= 0 && h < height && w < width) ?\n",
            "                   \n",
            "\u001b[01m\u001b[K./src/im2col_kernels.cu:1178:6:\u001b[m\u001b[K \u001b[01;35m\u001b[Kwarning: \u001b[m\u001b[K\"/*\" within comment [\u001b[01;35m\u001b[K-Wcomment\u001b[m\u001b[K]\n",
            "     //*((uint64_t *)(A_s + (local_i*lda + k) / 8)) = *((uint64_t *)(A + (i_cur*lda + k) / 8));    // weights\n",
            "       \n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/col2im_kernels.cu -o obj/col2im_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/blas_kernels.cu -o obj/blas_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/crop_layer_kernels.cu -o obj/crop_layer_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/dropout_layer_kernels.cu -o obj/dropout_layer_kernels.o\n",
            "./src/dropout_layer_kernels.cu(18): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(39): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(18): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(39): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(18): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(39): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(18): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(39): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(18): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "./src/dropout_layer_kernels.cu(39): warning: variable \"iteration_num\" was declared but never referenced\n",
            "\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/maxpool_layer_kernels.cu -o obj/maxpool_layer_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/network_kernels.cu -o obj/network_kernels.o\n",
            "nvcc -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] -gencode arch=compute_61,code=[sm_61,compute_61] -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN --compiler-options \"-Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include\" -c ./src/avgpool_layer_kernels.cu -o obj/avgpool_layer_kernels.o\n",
            "g++ -std=c++11 -Iinclude/ -I3rdparty/stb/include -DGPU -I/usr/local/cuda/include/ -DCUDNN -Wall -Wfatal-errors -Wno-unused-result -Wno-unknown-pragmas -fPIC -Ofast -DGPU -DCUDNN -I/usr/local/cudnn/include obj/image_opencv.o obj/http_stream.o obj/gemm.o obj/utils.o obj/dark_cuda.o obj/convolutional_layer.o obj/list.o obj/image.o obj/activations.o obj/im2col.o obj/col2im.o obj/blas.o obj/crop_layer.o obj/dropout_layer.o obj/maxpool_layer.o obj/softmax_layer.o obj/data.o obj/matrix.o obj/network.o obj/connected_layer.o obj/cost_layer.o obj/parser.o obj/option_list.o obj/darknet.o obj/detection_layer.o obj/captcha.o obj/route_layer.o obj/writing.o obj/box.o obj/nightmare.o obj/normalization_layer.o obj/avgpool_layer.o obj/coco.o obj/dice.o obj/yolo.o obj/detector.o obj/layer.o obj/compare.o obj/classifier.o obj/local_layer.o obj/swag.o obj/shortcut_layer.o obj/activation_layer.o obj/rnn_layer.o obj/gru_layer.o obj/rnn.o obj/rnn_vid.o obj/crnn_layer.o obj/demo.o obj/tag.o obj/cifar.o obj/go.o obj/batchnorm_layer.o obj/art.o obj/region_layer.o obj/reorg_layer.o obj/reorg_old_layer.o obj/super.o obj/voxel.o obj/tree.o obj/yolo_layer.o obj/upsample_layer.o obj/lstm_layer.o obj/conv_lstm_layer.o obj/scale_channels_layer.o obj/sam_layer.o obj/convolutional_kernels.o obj/activation_kernels.o obj/im2col_kernels.o obj/col2im_kernels.o obj/blas_kernels.o obj/crop_layer_kernels.o obj/dropout_layer_kernels.o obj/maxpool_layer_kernels.o obj/network_kernels.o obj/avgpool_layer_kernels.o -o darknet -lm -pthread -L/usr/local/cuda/lib64 -lcuda -lcudart -lcublas -lcurand -L/usr/local/cudnn/lib64 -lcudnn -lstdc++ \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "soiOKHQOVYWD",
        "colab_type": "code",
        "outputId": "28b62a93-9379-4ba0-ecb7-3e9c3a4188f1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "# if make command worked properly this should show \"usage: ./darknet <function>\"\"\n",
        "!./darknet"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "usage: ./darknet <function>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_AnCJQ4WKljp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!ls data/odonata/*.jpg >train.txt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ylkaPdObT45B",
        "colab_type": "code",
        "outputId": "79288d5f-f80b-4881-d920-782750a18e74",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 208
        }
      },
      "source": [
        "#!wget https://pjreddie.com/media/files/yolov3-openimages.weights"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2019-08-28 13:03:51--  https://pjreddie.com/media/files/yolov3-openimages.weights\n",
            "Resolving pjreddie.com (pjreddie.com)... 128.208.4.108\n",
            "Connecting to pjreddie.com (pjreddie.com)|128.208.4.108|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 259229388 (247M) [application/octet-stream]\n",
            "Saving to: ‘yolov3-openimages.weights’\n",
            "\n",
            "yolov3-openimages.w 100%[===================>] 247.22M  3.81MB/s    in 2m 54s  \n",
            "\n",
            "2019-08-28 13:06:46 (1.42 MB/s) - ‘yolov3-openimages.weights’ saved [259229388/259229388]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NXNjzJzKhPTB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!mv train.txt data/\n",
        "#!mv test.txt data/"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DRzd3sbqhxp7",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!mv backup/ backup_good/"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IRRMv2joMdTd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "%pycat data/odonata.data\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r1nlc2w-bjBE",
        "colab_type": "code",
        "outputId": "ee2f1f5e-20d4-42ee-e6db-5b64dcac7df0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#to start training with imagenet pretraind weight\n",
        "!./darknet detector train data/odonata.data yolov3-odonata.cfg darknet53.conv.74 -map >train.log"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from darknet53.conv.74...Done!\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "^C\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hlybGaiXMqtS",
        "colab_type": "code",
        "outputId": "d2c3a574-e1eb-4365-fbe6-5152302c9c89",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#to continue training from the weights of 1000th epoch\n",
        "!./darknet detector train data/odonata.data yolov3-odonata.cfg backup/yolov3-odonata_1000.weights >> train.log"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 24.92 MB \n",
            "Loading weights from backup/yolov3-odonata_1000.weights...Done!\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_2000.weights\n",
            "Saving weights to backup//yolov3-odonata_last.weights\n",
            "Saving weights to backup//yolov3-odonata_final.weights\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yaX-miuuShqo",
        "colab_type": "code",
        "outputId": "a5f91541-9f31-47fd-e96f-1c71560db184",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#prediction on a single image\n",
        "!./darknet detector test data/odonata.data yolov3-odonata-test.cfg backup/yolov3-odonata_final.weights 000000387352.jpg"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_final.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "000000387352.jpg: Predicted in 92.088000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZNZ5P-OmVfnI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#creating a list of untrained data for prediction\n",
        "#!ls ../predict/*.jpg > list.txt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K1YQFJaZcMTJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#!ls ../predict_final/*.jpg > list_f.txt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xru9WCLmfURi",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#view the created list\n",
        "%pycat list_f.txt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UOIRI01n_PC-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!mkdir results_aug30_imagenet"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kVzDuaDOfZcu",
        "colab_type": "code",
        "outputId": "c602f2cd-2e39-4c7b-b056-e74a9a993073",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "#run prediction on a list of images\n",
        "fname=open('list.txt')\n",
        "for i,line in enumerate(fname):\n",
        "  \n",
        "  x=line.rpartition(\"/\")[-1]\n",
        "  !./darknet detector test data/odonata.data yolov3-odonata-test.cfg backup/yolov3-odonata_best.weights $line \n",
        "  !cp predictions.jpg results_aug30/$x\n",
        "  if i==100:\n",
        "    break\n",
        "fname.close()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0462.jpg: Predicted in 87.331000 milli-seconds.\n",
            "dragonfly: 68%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0463.jpg: Predicted in 78.234000 milli-seconds.\n",
            "dragonfly: 84%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0464.jpg: Predicted in 72.185000 milli-seconds.\n",
            "dragonfly: 67%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0465.jpg: Predicted in 88.534000 milli-seconds.\n",
            "dragonfly: 40%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0466.jpg: Predicted in 86.601000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0467.jpg: Predicted in 77.832000 milli-seconds.\n",
            "dragonfly: 45%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0468.jpg: Predicted in 78.148000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0469.jpg: Predicted in 88.771000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0470.jpg: Predicted in 88.993000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0471.jpg: Predicted in 77.619000 milli-seconds.\n",
            "dragonfly: 64%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0472.jpg: Predicted in 72.056000 milli-seconds.\n",
            "dragonfly: 64%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0473.jpg: Predicted in 71.935000 milli-seconds.\n",
            "dragonfly: 66%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0474.jpg: Predicted in 89.010000 milli-seconds.\n",
            "dragonfly: 98%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0475.jpg: Predicted in 88.674000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0476.jpg: Predicted in 82.329000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0477.jpg: Predicted in 81.426000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0478.jpg: Predicted in 73.975000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0479.jpg: Predicted in 88.306000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0480.jpg: Predicted in 88.336000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0481.jpg: Predicted in 77.789000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0482.jpg: Predicted in 70.085000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0483.jpg: Predicted in 70.464000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0484.jpg: Predicted in 88.514000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0485.jpg: Predicted in 87.891000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0486.jpg: Predicted in 78.564000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0487.jpg: Predicted in 77.912000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0488.jpg: Predicted in 78.056000 milli-seconds.\n",
            "dragonfly: 98%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0489.jpg: Predicted in 88.272000 milli-seconds.\n",
            "dragonfly: 96%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0490.jpg: Predicted in 88.604000 milli-seconds.\n",
            "dragonfly: 86%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0491.jpg: Predicted in 87.975000 milli-seconds.\n",
            "dragonfly: 29%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0492.jpg: Predicted in 79.466000 milli-seconds.\n",
            "dragonfly: 40%\n",
            "dragonfly: 29%\n",
            "dragonfly: 28%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0493.jpg: Predicted in 79.925000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0494.jpg: Predicted in 80.436000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0495.jpg: Predicted in 88.216000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0496.jpg: Predicted in 80.707000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0497.jpg: Predicted in 73.423000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0498.jpg: Predicted in 72.548000 milli-seconds.\n",
            "dragonfly: 76%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0499.jpg: Predicted in 88.276000 milli-seconds.\n",
            "dragonfly: 78%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0500.jpg: Predicted in 81.120000 milli-seconds.\n",
            "dragonfly: 81%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0501.jpg: Predicted in 73.794000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0502.jpg: Predicted in 66.547000 milli-seconds.\n",
            "dragonfly: 66%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0503.jpg: Predicted in 66.680000 milli-seconds.\n",
            "dragonfly: 48%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0504.jpg: Predicted in 88.730000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0505.jpg: Predicted in 88.627000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0506.jpg: Predicted in 88.579000 milli-seconds.\n",
            "dragonfly: 74%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0507.jpg: Predicted in 78.255000 milli-seconds.\n",
            "dragonfly: 84%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0508.jpg: Predicted in 78.425000 milli-seconds.\n",
            "dragonfly: 89%\n",
            "dragonfly: 48%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0509.jpg: Predicted in 72.040000 milli-seconds.\n",
            "dragonfly: 56%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0510.jpg: Predicted in 87.766000 milli-seconds.\n",
            "dragonfly: 36%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0511.jpg: Predicted in 88.784000 milli-seconds.\n",
            "dragonfly: 74%\n",
            "dragonfly: 32%\n",
            "dragonfly: 66%\n",
            "dragonfly: 33%\n",
            "dragonfly: 52%\n",
            "dragonfly: 52%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0512.jpg: Predicted in 75.840000 milli-seconds.\n",
            "dragonfly: 43%\n",
            "dragonfly: 74%\n",
            "dragonfly: 40%\n",
            "dragonfly: 56%\n",
            "dragonfly: 81%\n",
            "dragonfly: 31%\n",
            "dragonfly: 72%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0513.jpg: Predicted in 71.994000 milli-seconds.\n",
            "dragonfly: 53%\n",
            "dragonfly: 40%\n",
            "dragonfly: 28%\n",
            "dragonfly: 42%\n",
            "dragonfly: 29%\n",
            "dragonfly: 59%\n",
            "dragonfly: 80%\n",
            "dragonfly: 79%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0514.jpg: Predicted in 70.278000 milli-seconds.\n",
            "dragonfly: 86%\n",
            "dragonfly: 41%\n",
            "dragonfly: 49%\n",
            "dragonfly: 26%\n",
            "dragonfly: 55%\n",
            "dragonfly: 56%\n",
            "dragonfly: 27%\n",
            "dragonfly: 32%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0515.jpg: Predicted in 88.254000 milli-seconds.\n",
            "dragonfly: 99%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0516.jpg: Predicted in 78.106000 milli-seconds.\n",
            "dragonfly: 100%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0517.jpg: Predicted in 77.941000 milli-seconds.\n",
            "dragonfly: 96%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0518.jpg: Predicted in 73.166000 milli-seconds.\n",
            "dragonfly: 97%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0519.jpg: Predicted in 88.138000 milli-seconds.\n",
            "dragonfly: 90%\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0520.jpg: Predicted in 77.778000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0521.jpg: Predicted in 77.641000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0522.jpg: Predicted in 77.444000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n",
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "../predict/DSC_0523.jpg: Predicted in 70.440000 milli-seconds.\n",
            "Not compiled with OpenCV, saving to predictions.png instead\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2cQmeiHZ9y42",
        "colab_type": "code",
        "outputId": "79a4d6af-1978-4f41-b3c2-5393475e2a2c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "ls predict_on_trained/ |wc -l"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "101\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GdioJCX1N47F",
        "colab_type": "code",
        "outputId": "c4979982-e0d7-4fa5-de2a-43bba2613fd3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 694
        }
      },
      "source": [
        "\n",
        "!ls -lt"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "total 179453\n",
            "-rw------- 1 root root    204954 Aug 16 06:55 predictions.jpg\n",
            "drwx------ 2 root root      4096 Aug 16 06:31 predict_on_trained\n",
            "-rw------- 1 root root   3511432 Aug 16 06:21 darknet\n",
            "-rw------- 1 root root      4992 Aug 16 06:18 Makefile\n",
            "drwx------ 2 root root      4096 Aug 15 20:30 results_aug16\n",
            "-rw------- 1 root root      5018 Aug 15 20:29 list.txt\n",
            "-rw------- 1 root root       860 Aug 15 20:25 bad.list\n",
            "-rw------- 1 root root  17356254 Aug 15 18:01 train.log\n",
            "-rw------- 1 root root      8329 Aug 15 17:35 yolov3-odonata.cfg\n",
            "drwx------ 2 root root      4096 Aug 15 10:31 backup\n",
            "drwx------ 2 root root      4096 Aug 15 10:05 weights_old\n",
            "drwx------ 2 root root      4096 Aug 10 18:14 results\n",
            "-rw------- 1 root root      8406 Aug 10 10:17 yolov3-odonata-test.cfg\n",
            "drwx------ 2 root root      4096 Aug  9 07:51 obj\n",
            "-rw------- 1 root root     20019 Aug  9 06:06 CMakeLists.txt\n",
            "-rw------- 1 root root     52260 Aug  9 06:06 README.md\n",
            "-rw------- 1 root root      8265 Aug  9 06:06 build.ps1\n",
            "-rw------- 1 root root      1950 Aug  9 06:06 build.sh\n",
            "-rw------- 1 root root      4010 Aug  9 06:06 darknet_video.py\n",
            "-rw------- 1 root root       108 Aug  9 06:06 video_v2.sh\n",
            "-rw------- 1 root root      1363 Aug  9 06:06 DarknetConfig.cmake.in\n",
            "-rw------- 1 root root     16517 Aug  9 06:06 darknet.py\n",
            "-rw------- 1 root root       108 Aug  9 06:06 image_yolov2.sh\n",
            "-rw------- 1 root root       345 Aug  9 06:06 json_mjpeg_streams.sh\n",
            "-rw------- 1 root root       515 Aug  9 06:06 LICENSE\n",
            "-rw------- 1 root root       159 Aug  9 06:06 net_cam_v3.sh\n",
            "-rw------- 1 root root       108 Aug  9 06:06 video_yolov3.sh\n",
            "-rw------- 1 root root      6344 Aug  9 06:06 appveyor.yml\n",
            "-rw------- 1 root root       110 Aug  9 06:06 image_yolov3.sh\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 3rdparty\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 build\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 cfg\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 cmake\n",
            "drwx------ 5 root root      4096 Aug  9 06:06 data\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 include\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 scripts\n",
            "drwx------ 2 root root      4096 Aug  9 06:06 src\n",
            "-rw------- 1 root root 162482580 Mar 25  2018 darknet53.conv.74\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uUgnpwIytz7B",
        "colab_type": "code",
        "outputId": "7ad4721b-3687-46c9-9345-46a576af58c0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 318
        }
      },
      "source": [
        "import sys\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "lines = []\n",
        "for line in open(\"train.log\"):\n",
        "    if \"avg\" in line:\n",
        "        lines.append(line)\n",
        "\n",
        "iterations = []\n",
        "avg_loss = []\n",
        "\n",
        "print('Retrieving data and plotting training loss graph...')\n",
        "for i in range(len(lines)):\n",
        "    lineParts = lines[i].split(',')\n",
        "    iterations.append(int(lineParts[0].split(':')[0]))\n",
        "    avg_loss.append(float(lineParts[1].split()[0]))\n",
        "\n",
        "fig = plt.figure()\n",
        "for i in range(0, len(lines)):\n",
        "    plt.plot(iterations[i:i+2], avg_loss[i:i+2], 'r.-')\n",
        "\n",
        "plt.xlabel('Batch Number')\n",
        "plt.ylabel('Avg Loss')\n",
        "fig.savefig('training_loss_plot.png', dpi=1000)\n",
        "\n",
        "print('Done! Plot saved as training_loss_plot.png')\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Retrieving data and plotting training loss graph...\n",
            "Done! Plot saved as training_loss_plot.png\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEKCAYAAADq59mMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6TE4IjxUxpp-",
        "colab_type": "code",
        "outputId": "59da25a8-8261-48a8-afd5-5274c8e3b4ae",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "!./darknet detector map data/odonata.data yolov3-odonata.cfg backup/yolov3-odonata_3000.weights"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_3000.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "\n",
            " calculation mAP (mean average precision)...\n",
            "148\n",
            " detections_count = 192, unique_truth_count = 150  \n",
            "class_id = 0, name = dragonfly, ap = 96.34%   \t (TP = 143, FP = 7) \n",
            "\n",
            " for conf_thresh = 0.25, precision = 0.95, recall = 0.95, F1-score = 0.95 \n",
            " for conf_thresh = 0.25, TP = 143, FP = 7, FN = 7, average IoU = 76.72 % \n",
            "\n",
            " IoU threshold = 50 %, used Area-Under-Curve for each unique Recall \n",
            " mean average precision (mAP@0.50) = 0.963353, or 96.34 % \n",
            "Total Detection Time: 82.000000 Seconds\n",
            "\n",
            "Set -points flag:\n",
            " `-points 101` for MS COCO \n",
            " `-points 11` for PascalVOC 2007 (uncomment `difficult` in voc.data) \n",
            " `-points 0` (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CQgyv-1dYPiE",
        "colab_type": "code",
        "outputId": "2c6a7fe7-0f43-4f09-e4ca-3dbee8c83c69",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "!./darknet detector map data/odonata.data yolov3-odonata.cfg backup/yolov3-odonata_final.weights\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_final.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "\n",
            " calculation mAP (mean average precision)...\n",
            "148\n",
            " detections_count = 162, unique_truth_count = 150  \n",
            "class_id = 0, name = dragonfly, ap = 97.26%   \t (TP = 146, FP = 3) \n",
            "\n",
            " for conf_thresh = 0.25, precision = 0.98, recall = 0.97, F1-score = 0.98 \n",
            " for conf_thresh = 0.25, TP = 146, FP = 3, FN = 4, average IoU = 82.33 % \n",
            "\n",
            " IoU threshold = 50 %, used Area-Under-Curve for each unique Recall \n",
            " mean average precision (mAP@0.50) = 0.972613, or 97.26 % \n",
            "Total Detection Time: 14.000000 Seconds\n",
            "\n",
            "Set -points flag:\n",
            " `-points 101` for MS COCO \n",
            " `-points 11` for PascalVOC 2007 (uncomment `difficult` in voc.data) \n",
            " `-points 0` (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D7jgFmm7t5Jb",
        "colab_type": "code",
        "outputId": "413f0a2a-472e-4af3-94ce-7385380e494b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "!./darknet detector map data/odonata.data yolov3-odonata.cfg backup/yolov3-odonata_2000.weights"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_2000.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "\n",
            " calculation mAP (mean average precision)...\n",
            "148\n",
            " detections_count = 194, unique_truth_count = 150  \n",
            "class_id = 0, name = dragonfly, ap = 97.15%   \t (TP = 144, FP = 3) \n",
            "\n",
            " for conf_thresh = 0.25, precision = 0.98, recall = 0.96, F1-score = 0.97 \n",
            " for conf_thresh = 0.25, TP = 144, FP = 3, FN = 6, average IoU = 79.35 % \n",
            "\n",
            " IoU threshold = 50 %, used Area-Under-Curve for each unique Recall \n",
            " mean average precision (mAP@0.50) = 0.971534, or 97.15 % \n",
            "Total Detection Time: 14.000000 Seconds\n",
            "\n",
            "Set -points flag:\n",
            " `-points 101` for MS COCO \n",
            " `-points 11` for PascalVOC 2007 (uncomment `difficult` in voc.data) \n",
            " `-points 0` (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6umKTtjFWyf0",
        "colab_type": "code",
        "outputId": "ff866232-ab3b-496c-98ce-abd25358cf5d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "!./darknet detector map data/odonata.data yolov3-odonata.cfg backup/yolov3-odonata_1000.weights"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_1000.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "\n",
            " calculation mAP (mean average precision)...\n",
            "148\n",
            " detections_count = 372, unique_truth_count = 150  \n",
            "class_id = 0, name = dragonfly, ap = 63.31%   \t (TP = 96, FP = 36) \n",
            "\n",
            " for conf_thresh = 0.25, precision = 0.73, recall = 0.64, F1-score = 0.68 \n",
            " for conf_thresh = 0.25, TP = 96, FP = 36, FN = 54, average IoU = 46.33 % \n",
            "\n",
            " IoU threshold = 50 %, used Area-Under-Curve for each unique Recall \n",
            " mean average precision (mAP@0.50) = 0.633117, or 63.31 % \n",
            "Total Detection Time: 59.000000 Seconds\n",
            "\n",
            "Set -points flag:\n",
            " `-points 101` for MS COCO \n",
            " `-points 11` for PascalVOC 2007 (uncomment `difficult` in voc.data) \n",
            " `-points 0` (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NYJUQ22GXNJl",
        "colab_type": "code",
        "outputId": "e12dc33b-c347-4be3-9601-c68068f2aec7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "!./darknet detector map data/odonata.data yolov3-odonata.cfg backup/yolov3-odonata_best.weights"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   layer   filters  size/strd(dil)      input                output\n",
            "   0 conv     32       3 x 3/ 1    416 x 416 x   3 ->  416 x 416 x  32 0.299 BF\n",
            "   1 conv     64       3 x 3/ 2    416 x 416 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   2 conv     32       1 x 1/ 1    208 x 208 x  64 ->  208 x 208 x  32 0.177 BF\n",
            "   3 conv     64       3 x 3/ 1    208 x 208 x  32 ->  208 x 208 x  64 1.595 BF\n",
            "   4 Shortcut Layer: 1\n",
            "   5 conv    128       3 x 3/ 2    208 x 208 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   6 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "   7 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "   8 Shortcut Layer: 5\n",
            "   9 conv     64       1 x 1/ 1    104 x 104 x 128 ->  104 x 104 x  64 0.177 BF\n",
            "  10 conv    128       3 x 3/ 1    104 x 104 x  64 ->  104 x 104 x 128 1.595 BF\n",
            "  11 Shortcut Layer: 8\n",
            "  12 conv    256       3 x 3/ 2    104 x 104 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  13 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  14 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  15 Shortcut Layer: 12\n",
            "  16 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  17 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  18 Shortcut Layer: 15\n",
            "  19 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  20 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  21 Shortcut Layer: 18\n",
            "  22 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  23 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  24 Shortcut Layer: 21\n",
            "  25 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  26 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  27 Shortcut Layer: 24\n",
            "  28 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  29 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  30 Shortcut Layer: 27\n",
            "  31 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  32 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  33 Shortcut Layer: 30\n",
            "  34 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            "  35 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            "  36 Shortcut Layer: 33\n",
            "  37 conv    512       3 x 3/ 2     52 x  52 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  38 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  39 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  40 Shortcut Layer: 37\n",
            "  41 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  42 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  43 Shortcut Layer: 40\n",
            "  44 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  45 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  46 Shortcut Layer: 43\n",
            "  47 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  48 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  49 Shortcut Layer: 46\n",
            "  50 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  51 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  52 Shortcut Layer: 49\n",
            "  53 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  54 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  55 Shortcut Layer: 52\n",
            "  56 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  57 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  58 Shortcut Layer: 55\n",
            "  59 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  60 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  61 Shortcut Layer: 58\n",
            "  62 conv   1024       3 x 3/ 2     26 x  26 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  63 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  64 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  65 Shortcut Layer: 62\n",
            "  66 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  67 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  68 Shortcut Layer: 65\n",
            "  69 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  70 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  71 Shortcut Layer: 68\n",
            "  72 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  73 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  74 Shortcut Layer: 71\n",
            "  75 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  76 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  77 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  78 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  79 conv    512       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x 512 0.177 BF\n",
            "  80 conv   1024       3 x 3/ 1     13 x  13 x 512 ->   13 x  13 x1024 1.595 BF\n",
            "  81 conv     18       1 x 1/ 1     13 x  13 x1024 ->   13 x  13 x  18 0.006 BF\n",
            "  82 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  83 route  79\n",
            "  84 conv    256       1 x 1/ 1     13 x  13 x 512 ->   13 x  13 x 256 0.044 BF\n",
            "  85 upsample                 2x    13 x  13 x 256 ->   26 x  26 x 256\n",
            "  86 route  85 61\n",
            "  87 conv    256       1 x 1/ 1     26 x  26 x 768 ->   26 x  26 x 256 0.266 BF\n",
            "  88 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  89 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  90 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  91 conv    256       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x 256 0.177 BF\n",
            "  92 conv    512       3 x 3/ 1     26 x  26 x 256 ->   26 x  26 x 512 1.595 BF\n",
            "  93 conv     18       1 x 1/ 1     26 x  26 x 512 ->   26 x  26 x  18 0.012 BF\n",
            "  94 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "  95 route  91\n",
            "  96 conv    128       1 x 1/ 1     26 x  26 x 256 ->   26 x  26 x 128 0.044 BF\n",
            "  97 upsample                 2x    26 x  26 x 128 ->   52 x  52 x 128\n",
            "  98 route  97 36\n",
            "  99 conv    128       1 x 1/ 1     52 x  52 x 384 ->   52 x  52 x 128 0.266 BF\n",
            " 100 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 101 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 102 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 103 conv    128       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x 128 0.177 BF\n",
            " 104 conv    256       3 x 3/ 1     52 x  52 x 128 ->   52 x  52 x 256 1.595 BF\n",
            " 105 conv     18       1 x 1/ 1     52 x  52 x 256 ->   52 x  52 x  18 0.025 BF\n",
            " 106 yolo\n",
            "[yolo] params: iou loss: mse, iou_norm: 0.75, cls_norm: 1.00, scale_x_y: 1.00\n",
            "Total BFLOPS 65.290 \n",
            " Allocate additional workspace_size = 12.46 MB \n",
            "Loading weights from backup/yolov3-odonata_best.weights...\n",
            " seen 64 \n",
            "Done!\n",
            "\n",
            " calculation mAP (mean average precision)...\n",
            "128\n",
            " detections_count = 271, unique_truth_count = 129  \n",
            "class_id = 0, name = dragonfly, ap = 84.69%   \t (TP = 105, FP = 24) \n",
            "\n",
            " for conf_thresh = 0.25, precision = 0.81, recall = 0.81, F1-score = 0.81 \n",
            " for conf_thresh = 0.25, TP = 105, FP = 24, FN = 24, average IoU = 56.06 % \n",
            "\n",
            " IoU threshold = 50 %, used Area-Under-Curve for each unique Recall \n",
            " mean average precision (mAP@0.50) = 0.846918, or 84.69 % \n",
            "Total Detection Time: 61.000000 Seconds\n",
            "\n",
            "Set -points flag:\n",
            " `-points 101` for MS COCO \n",
            " `-points 11` for PascalVOC 2007 (uncomment `difficult` in voc.data) \n",
            " `-points 0` (AUC) for ImageNet, PascalVOC 2010-2012, your custom dataset\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7VcWUvmqNZ_t",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}