Speech_Emotion _Recognition_v3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Speech Emotion Recognition Version 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import librosa\n",
    "import librosa.display\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "import os\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import sys\n",
    "import glob \n",
    "import warnings\n",
    "import keras\n",
    "from keras.preprocessing import sequence\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Embedding\n",
    "from keras.layers import LSTM\n",
    "from keras.preprocessing.text import Tokenizer\n",
    "from keras.preprocessing.sequence import pad_sequences\n",
    "from keras.utils import to_categorical\n",
    "from keras.layers import Input, Flatten, Dropout, Activation\n",
    "from keras.layers import Conv2D, Conv1D, MaxPooling2D, AveragePooling1D\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.models import Model\n",
    "from keras import regularizers\n",
    "from keras.optimizers import Adam\n",
    "from keras.regularizers import l2\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.utils import shuffle\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from tensorflow.contrib.tensor_forest.python import tensor_forest\n",
    "from tensorflow.python.ops import resources\n",
    "from tqdm import tqdm\n",
    "from matplotlib.pyplot import specgram\n",
    "from typing import Tuple\n",
    "# import scipy.io.wavfile as wav\n",
    "# from speechpy.feature import mfcc\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataList = os.listdir('data2/')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "classLabels = ('Angry', 'Fear', 'Disgust', 'Happy', 'Sad', 'Surprised', 'Neutral')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = []\n",
    "labels = []\n",
    "names = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1440/1440 [02:18<00:00, 11.01it/s]\n"
     ]
    }
   ],
   "source": [
    "for number, path in enumerate(tqdm(dataList)):\n",
    "    X, sample_rate = librosa.load('data2/'+path, res_type='kaiser_best',duration=2.5,sr=22050*2,offset=0.5)\n",
    "    sample_rate = np.array(sample_rate)\n",
    "    mfccs = librosa.feature.mfcc(y=X, sr=sample_rate, n_mfcc=39)\n",
    "    feature = mfccs\n",
    "    data.append(feature)\n",
    "    \n",
    "    if path[6:8] == '01' or path[0:1] == 'n':\n",
    "        labels.append(6)\n",
    "    if path[6:8] == '02':\n",
    "        labels.append(6)\n",
    "    if path[6:8] == '03' or path[0:1] == 'h':\n",
    "        labels.append(3)\n",
    "    if path[6:8] == '04' or path[0:2] == 'sa':\n",
    "        labels.append(4)\n",
    "    if path[6:8] == '05' or path[0:1] == 'a':\n",
    "        labels.append(0)\n",
    "    if path[6:8] == '06' or path[0:1] == 'f':\n",
    "        labels.append(1)\n",
    "    if path[6:8] == '07' or path[0:1] == 'd':\n",
    "        labels.append(2)\n",
    "    if path[6:8] == '08' or path[0:2] == 'su':\n",
    "        labels.append(5)\n",
    "    \n",
    "    \n",
    "    names.append(path)\n",
    "#     break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(39, 216)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA = np.zeros((1440,39,216))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "  0%|          | 0/1439 [00:00<?, ?it/s]\n",
      "\n",
      "  1%|          | 13/1439 [00:00<00:11, 119.33it/s]\n",
      "\n",
      "  2%|▏         | 26/1439 [00:00<00:11, 118.10it/s]\n",
      "\n",
      "  3%|▎         | 39/1439 [00:00<00:11, 119.52it/s]\n",
      "\n",
      "  4%|▎         | 51/1439 [00:00<00:11, 117.73it/s]\n",
      "\n",
      "  4%|▍         | 64/1439 [00:00<00:11, 119.66it/s]\n",
      "\n",
      "  5%|▌         | 78/1439 [00:00<00:11, 121.08it/s]\n",
      "\n",
      "  6%|▋         | 91/1439 [00:00<00:11, 119.36it/s]\n",
      "\n",
      "  7%|▋         | 104/1439 [00:00<00:11, 117.95it/s]\n",
      "\n",
      "  8%|▊         | 117/1439 [00:00<00:11, 118.48it/s]\n",
      "\n",
      "  9%|▉         | 130/1439 [00:01<00:10, 119.05it/s]\n",
      "\n",
      " 10%|▉         | 143/1439 [00:01<00:10, 119.43it/s]\n",
      "\n",
      " 11%|█         | 156/1439 [00:01<00:10, 118.98it/s]\n",
      "\n",
      " 12%|█▏        | 168/1439 [00:01<00:10, 116.75it/s]\n",
      "\n",
      " 13%|█▎        | 180/1439 [00:01<00:10, 116.06it/s]\n",
      "\n",
      " 13%|█▎        | 194/1439 [00:01<00:10, 118.97it/s]\n",
      "\n",
      " 14%|█▍        | 207/1439 [00:01<00:10, 118.28it/s]\n",
      "\n",
      " 15%|█▌        | 219/1439 [00:01<00:10, 111.95it/s]\n",
      "\n",
      " 16%|█▌        | 231/1439 [00:01<00:10, 113.04it/s]\n",
      "\n",
      " 17%|█▋        | 245/1439 [00:02<00:10, 116.41it/s]\n",
      "\n",
      " 18%|█▊        | 257/1439 [00:02<00:10, 117.16it/s]\n",
      "\n",
      " 19%|█▊        | 269/1439 [00:02<00:09, 117.94it/s]\n",
      "\n",
      " 20%|█▉        | 281/1439 [00:02<00:09, 118.34it/s]\n",
      "\n",
      " 20%|██        | 293/1439 [00:02<00:09, 118.70it/s]\n",
      "\n",
      " 21%|██▏       | 306/1439 [00:02<00:09, 117.92it/s]\n",
      "\n",
      " 22%|██▏       | 319/1439 [00:02<00:09, 118.70it/s]\n",
      "\n",
      " 23%|██▎       | 333/1439 [00:02<00:09, 120.85it/s]\n",
      "\n",
      " 24%|██▍       | 346/1439 [00:02<00:09, 118.42it/s]\n",
      "\n",
      " 25%|██▌       | 360/1439 [00:03<00:09, 119.24it/s]\n",
      "\n",
      " 26%|██▌       | 373/1439 [00:03<00:09, 117.92it/s]\n",
      "\n",
      " 27%|██▋       | 387/1439 [00:03<00:08, 119.58it/s]\n",
      "\n",
      " 28%|██▊       | 399/1439 [00:03<00:08, 118.33it/s]\n",
      "\n",
      " 29%|██▊       | 413/1439 [00:03<00:08, 119.31it/s]\n",
      "\n",
      " 30%|██▉       | 427/1439 [00:03<00:08, 120.11it/s]\n",
      "\n",
      " 31%|███       | 441/1439 [00:03<00:08, 120.42it/s]\n",
      "\n",
      " 32%|███▏      | 454/1439 [00:03<00:08, 118.61it/s]\n",
      "\n",
      " 32%|███▏      | 467/1439 [00:03<00:08, 119.41it/s]\n",
      "\n",
      " 33%|███▎      | 481/1439 [00:04<00:07, 119.91it/s]\n",
      "\n",
      " 34%|███▍      | 494/1439 [00:04<00:08, 118.01it/s]\n",
      "\n",
      " 35%|███▌      | 508/1439 [00:04<00:07, 119.79it/s]\n",
      "\n",
      " 36%|███▌      | 520/1439 [00:04<00:07, 119.21it/s]\n",
      "\n",
      " 37%|███▋      | 532/1439 [00:04<00:07, 115.23it/s]\n",
      "\n",
      " 38%|███▊      | 546/1439 [00:04<00:07, 118.92it/s]\n",
      "\n",
      " 39%|███▉      | 558/1439 [00:04<00:07, 118.76it/s]\n",
      "\n",
      " 40%|███▉      | 570/1439 [00:04<00:07, 118.92it/s]\n",
      "\n",
      " 41%|████      | 583/1439 [00:04<00:07, 118.09it/s]\n",
      "\n",
      " 41%|████▏     | 596/1439 [00:05<00:07, 118.30it/s]\n",
      "\n",
      " 42%|████▏     | 609/1439 [00:05<00:07, 117.32it/s]\n",
      "\n",
      " 43%|████▎     | 622/1439 [00:05<00:06, 117.88it/s]\n",
      "\n",
      " 44%|████▍     | 635/1439 [00:05<00:06, 118.87it/s]\n",
      "\n",
      " 45%|████▍     | 647/1439 [00:05<00:06, 118.25it/s]\n",
      "\n",
      " 46%|████▌     | 661/1439 [00:05<00:06, 119.42it/s]\n",
      "\n",
      " 47%|████▋     | 674/1439 [00:05<00:06, 119.56it/s]\n",
      "\n",
      " 48%|████▊     | 687/1439 [00:05<00:06, 120.02it/s]\n",
      "\n",
      " 49%|████▊     | 700/1439 [00:05<00:06, 122.32it/s]\n",
      "\n",
      " 50%|████▉     | 713/1439 [00:06<00:06, 120.44it/s]\n",
      "\n",
      " 50%|█████     | 726/1439 [00:06<00:06, 118.01it/s]\n",
      "\n",
      " 51%|█████▏    | 738/1439 [00:06<00:05, 117.30it/s]\n",
      "\n",
      " 52%|█████▏    | 751/1439 [00:06<00:05, 120.09it/s]\n",
      "\n",
      " 53%|█████▎    | 764/1439 [00:06<00:05, 118.30it/s]\n",
      "\n",
      " 54%|█████▍    | 777/1439 [00:06<00:05, 119.63it/s]\n",
      "\n",
      " 55%|█████▍    | 790/1439 [00:06<00:05, 121.27it/s]\n",
      "\n",
      " 56%|█████▌    | 803/1439 [00:06<00:05, 121.34it/s]\n",
      "\n",
      " 57%|█████▋    | 816/1439 [00:06<00:05, 121.25it/s]\n",
      "\n",
      " 58%|█████▊    | 829/1439 [00:06<00:05, 121.13it/s]\n",
      "\n",
      " 59%|█████▊    | 842/1439 [00:07<00:04, 121.18it/s]\n",
      "\n",
      " 59%|█████▉    | 855/1439 [00:07<00:04, 119.05it/s]\n",
      "\n",
      " 60%|██████    | 867/1439 [00:07<00:04, 117.06it/s]\n",
      "\n",
      " 61%|██████    | 880/1439 [00:07<00:04, 119.70it/s]\n",
      "\n",
      " 62%|██████▏   | 893/1439 [00:07<00:04, 119.12it/s]\n",
      "\n",
      " 63%|██████▎   | 906/1439 [00:07<00:04, 120.29it/s]\n",
      "\n",
      " 64%|██████▍   | 919/1439 [00:07<00:04, 121.96it/s]\n",
      "\n",
      " 65%|██████▍   | 932/1439 [00:07<00:04, 120.11it/s]\n",
      "\n",
      " 66%|██████▌   | 945/1439 [00:07<00:04, 116.26it/s]\n",
      "\n",
      " 67%|██████▋   | 958/1439 [00:08<00:04, 118.05it/s]\n",
      "\n",
      " 67%|██████▋   | 970/1439 [00:08<00:04, 115.76it/s]\n",
      "\n",
      " 68%|██████▊   | 983/1439 [00:08<00:03, 118.10it/s]\n",
      "\n",
      " 69%|██████▉   | 996/1439 [00:08<00:03, 117.68it/s]\n",
      "\n",
      " 70%|███████   | 1008/1439 [00:08<00:03, 116.88it/s]\n",
      "\n",
      " 71%|███████   | 1020/1439 [00:08<00:03, 116.67it/s]\n",
      "\n",
      " 72%|███████▏  | 1034/1439 [00:08<00:03, 118.81it/s]\n",
      "\n",
      " 73%|███████▎  | 1046/1439 [00:08<00:03, 117.47it/s]\n",
      "\n",
      " 74%|███████▎  | 1058/1439 [00:08<00:03, 116.02it/s]\n",
      "\n",
      " 74%|███████▍  | 1071/1439 [00:09<00:03, 116.23it/s]\n",
      "\n",
      " 75%|███████▌  | 1084/1439 [00:09<00:03, 115.29it/s]\n",
      "\n",
      " 76%|███████▌  | 1096/1439 [00:09<00:02, 115.62it/s]\n",
      "\n",
      " 77%|███████▋  | 1108/1439 [00:09<00:02, 116.74it/s]\n",
      "\n",
      " 78%|███████▊  | 1121/1439 [00:09<00:02, 115.58it/s]\n",
      "\n",
      " 79%|███████▉  | 1134/1439 [00:09<00:02, 116.86it/s]\n",
      "\n",
      " 80%|███████▉  | 1147/1439 [00:09<00:02, 115.73it/s]\n",
      "\n",
      " 81%|████████  | 1161/1439 [00:09<00:02, 117.33it/s]\n",
      "\n",
      " 82%|████████▏ | 1173/1439 [00:09<00:02, 116.27it/s]\n",
      "\n",
      " 82%|████████▏ | 1186/1439 [00:10<00:02, 115.50it/s]\n",
      "\n",
      " 83%|████████▎ | 1199/1439 [00:10<00:02, 117.62it/s]\n",
      "\n",
      " 84%|████████▍ | 1211/1439 [00:10<00:01, 114.99it/s]\n",
      "\n",
      " 85%|████████▍ | 1223/1439 [00:10<00:01, 115.99it/s]\n",
      "\n",
      " 86%|████████▌ | 1235/1439 [00:10<00:01, 112.64it/s]\n",
      "\n",
      " 87%|████████▋ | 1247/1439 [00:10<00:01, 113.09it/s]\n",
      "\n",
      " 88%|████████▊ | 1260/1439 [00:10<00:01, 116.83it/s]\n",
      "\n",
      " 88%|████████▊ | 1272/1439 [00:10<00:01, 114.80it/s]\n",
      "\n",
      " 89%|████████▉ | 1284/1439 [00:10<00:01, 115.04it/s]\n",
      "\n",
      " 90%|█████████ | 1297/1439 [00:10<00:01, 116.07it/s]\n",
      "\n",
      " 91%|█████████ | 1310/1439 [00:11<00:01, 117.51it/s]\n",
      "\n",
      " 92%|█████████▏| 1323/1439 [00:11<00:00, 116.18it/s]\n",
      "\n",
      " 93%|█████████▎| 1336/1439 [00:11<00:00, 115.10it/s]\n",
      "\n",
      " 94%|█████████▎| 1348/1439 [00:11<00:00, 113.70it/s]\n",
      "\n",
      " 95%|█████████▍| 1360/1439 [00:11<00:00, 114.95it/s]\n",
      "\n",
      " 95%|█████████▌| 1374/1439 [00:11<00:00, 117.03it/s]\n",
      "\n",
      " 96%|█████████▋| 1387/1439 [00:11<00:00, 118.17it/s]\n",
      "\n",
      " 97%|█████████▋| 1400/1439 [00:11<00:00, 120.04it/s]\n",
      "\n",
      " 98%|█████████▊| 1413/1439 [00:11<00:00, 118.69it/s]\n",
      "\n",
      " 99%|█████████▉| 1425/1439 [00:12<00:00, 116.36it/s]\n",
      "\n",
      "100%|█████████▉| 1437/1439 [00:12<00:00, 116.53it/s]\n",
      "\n",
      "100%|██████████| 1439/1439 [00:12<00:00, 117.96it/s]"
     ]
    }
   ],
   "source": [
    "for i in tqdm(range(0, 1439)):\n",
    "    for j in range(0, 39):\n",
    "        for k in range(0, 216):\n",
    "#             print(i)\n",
    "#             print(j)\n",
    "#             print(k)\n",
    "            temp = data[i]\n",
    "            temp2 = temp[j]\n",
    "            if (k >= len(temp2)):\n",
    "                continue\n",
    "            else:\n",
    "                temp3 = temp2[k]\n",
    "            DATA[i][j][k] = float(temp3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1440, 39, 216)"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DATA = np.array(DATA)\n",
    "DATA.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1440"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels = np.array(labels)\n",
    "len(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(DATA, labels, test_size=0.3, random_state=42)\n",
    "\n",
    "X_train = np.array(X_train)\n",
    "X_test = np.array(X_test)\n",
    "\n",
    "Y_train = np.array(Y_train)\n",
    "Y_test = np.array(Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "numLabels = len(classLabels)\n",
    "\n",
    "Y_train = to_categorical(Y_train)\n",
    "Y_test = to_categorical(Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "in_shape = X_train[0].shape\n",
    "X_train = X_train.reshape(X_train.shape[0], in_shape[0], in_shape[1], 1)\n",
    "X_test = X_test.reshape(X_test.shape[0], in_shape[0], in_shape[1], 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(39, 216)"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "in_shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Logging before flag parsing goes to stderr.\n",
      "W0513 01:11:39.018074 18248 deprecation_wrapper.py:119] From C:\\Users\\samya\\Anaconda3\\lib\\site-packages\\keras\\backend\\tensorflow_backend.py:4070: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv2D(8, (13, 13), input_shape=(in_shape[0], in_shape[1], 1)))\n",
    "model.add(BatchNormalization(axis=-1))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Conv2D(8, (13, 13)))\n",
    "model.add(BatchNormalization(axis=-1))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 1)))\n",
    "model.add(Conv2D(8, (3, 3)))\n",
    "model.add(BatchNormalization(axis=-1))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Conv2D(8, (1, 1)))\n",
    "model.add(BatchNormalization(axis=-1))\n",
    "model.add(Activation('relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 1)))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(64))\n",
    "model.add(BatchNormalization())\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(numLabels, activation = 'softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 27, 204, 8)        1360      \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 27, 204, 8)        32        \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 27, 204, 8)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 15, 192, 8)        10824     \n",
      "_________________________________________________________________\n",
      "batch_normalization_2 (Batch (None, 15, 192, 8)        32        \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 15, 192, 8)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 7, 192, 8)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 5, 190, 8)         584       \n",
      "_________________________________________________________________\n",
      "batch_normalization_3 (Batch (None, 5, 190, 8)         32        \n",
      "_________________________________________________________________\n",
      "activation_3 (Activation)    (None, 5, 190, 8)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 5, 190, 8)         72        \n",
      "_________________________________________________________________\n",
      "batch_normalization_4 (Batch (None, 5, 190, 8)         32        \n",
      "_________________________________________________________________\n",
      "activation_4 (Activation)    (None, 5, 190, 8)         0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 2, 190, 8)         0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 3040)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 64)                194624    \n",
      "_________________________________________________________________\n",
      "batch_normalization_5 (Batch (None, 64)                256       \n",
      "_________________________________________________________________\n",
      "activation_5 (Activation)    (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 7)                 455       \n",
      "=================================================================\n",
      "Total params: 208,303\n",
      "Trainable params: 208,111\n",
      "Non-trainable params: 192\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='binary_crossentropy', optimizer='adam',\n",
    "                           metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "1008/1008 [==============================] - ETA: 32s - loss: 0.3655 - accuracy: 0.870 - ETA: 24s - loss: 0.3474 - accuracy: 0.875 - ETA: 21s - loss: 0.3309 - accuracy: 0.873 - ETA: 19s - loss: 0.3211 - accuracy: 0.879 - ETA: 18s - loss: 0.3152 - accuracy: 0.878 - ETA: 17s - loss: 0.3092 - accuracy: 0.882 - ETA: 16s - loss: 0.3071 - accuracy: 0.883 - ETA: 15s - loss: 0.3008 - accuracy: 0.883 - ETA: 14s - loss: 0.3001 - accuracy: 0.885 - ETA: 15s - loss: 0.2973 - accuracy: 0.885 - ETA: 14s - loss: 0.2937 - accuracy: 0.888 - ETA: 13s - loss: 0.2976 - accuracy: 0.885 - ETA: 12s - loss: 0.2970 - accuracy: 0.885 - ETA: 11s - loss: 0.2963 - accuracy: 0.884 - ETA: 10s - loss: 0.2953 - accuracy: 0.883 - ETA: 10s - loss: 0.2923 - accuracy: 0.885 - ETA: 9s - loss: 0.2914 - accuracy: 0.886 - ETA: 8s - loss: 0.2934 - accuracy: 0.88 - ETA: 8s - loss: 0.2937 - accuracy: 0.88 - ETA: 7s - loss: 0.2911 - accuracy: 0.88 - ETA: 7s - loss: 0.2900 - accuracy: 0.88 - ETA: 6s - loss: 0.2887 - accuracy: 0.88 - ETA: 5s - loss: 0.2884 - accuracy: 0.88 - ETA: 5s - loss: 0.2861 - accuracy: 0.88 - ETA: 4s - loss: 0.2852 - accuracy: 0.88 - ETA: 3s - loss: 0.2848 - accuracy: 0.88 - ETA: 2s - loss: 0.2849 - accuracy: 0.88 - ETA: 2s - loss: 0.2828 - accuracy: 0.88 - ETA: 1s - loss: 0.2833 - accuracy: 0.88 - ETA: 1s - loss: 0.2826 - accuracy: 0.88 - ETA: 0s - loss: 0.2814 - accuracy: 0.88 - 21s 21ms/step - loss: 0.2811 - accuracy: 0.8876\n",
      "Epoch 2/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.2392 - accuracy: 0.897 - ETA: 16s - loss: 0.2141 - accuracy: 0.912 - ETA: 15s - loss: 0.2271 - accuracy: 0.909 - ETA: 15s - loss: 0.2198 - accuracy: 0.909 - ETA: 14s - loss: 0.2225 - accuracy: 0.908 - ETA: 16s - loss: 0.2204 - accuracy: 0.910 - ETA: 15s - loss: 0.2145 - accuracy: 0.914 - ETA: 15s - loss: 0.2145 - accuracy: 0.913 - ETA: 14s - loss: 0.2151 - accuracy: 0.914 - ETA: 13s - loss: 0.2199 - accuracy: 0.910 - ETA: 12s - loss: 0.2219 - accuracy: 0.908 - ETA: 11s - loss: 0.2229 - accuracy: 0.907 - ETA: 11s - loss: 0.2214 - accuracy: 0.909 - ETA: 10s - loss: 0.2167 - accuracy: 0.912 - ETA: 10s - loss: 0.2165 - accuracy: 0.912 - ETA: 9s - loss: 0.2134 - accuracy: 0.914 - ETA: 9s - loss: 0.2149 - accuracy: 0.91 - ETA: 8s - loss: 0.2169 - accuracy: 0.91 - ETA: 7s - loss: 0.2173 - accuracy: 0.91 - ETA: 7s - loss: 0.2170 - accuracy: 0.91 - ETA: 6s - loss: 0.2185 - accuracy: 0.91 - ETA: 5s - loss: 0.2183 - accuracy: 0.91 - ETA: 5s - loss: 0.2205 - accuracy: 0.91 - ETA: 4s - loss: 0.2194 - accuracy: 0.91 - ETA: 4s - loss: 0.2205 - accuracy: 0.91 - ETA: 3s - loss: 0.2206 - accuracy: 0.91 - ETA: 2s - loss: 0.2207 - accuracy: 0.91 - ETA: 2s - loss: 0.2216 - accuracy: 0.91 - ETA: 1s - loss: 0.2221 - accuracy: 0.91 - ETA: 0s - loss: 0.2221 - accuracy: 0.91 - ETA: 0s - loss: 0.2213 - accuracy: 0.91 - 19s 19ms/step - loss: 0.2220 - accuracy: 0.9111\n",
      "Epoch 3/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.1726 - accuracy: 0.933 - ETA: 16s - loss: 0.1834 - accuracy: 0.928 - ETA: 21s - loss: 0.1694 - accuracy: 0.937 - ETA: 19s - loss: 0.1744 - accuracy: 0.933 - ETA: 18s - loss: 0.1712 - accuracy: 0.936 - ETA: 17s - loss: 0.1670 - accuracy: 0.936 - ETA: 16s - loss: 0.1651 - accuracy: 0.938 - ETA: 15s - loss: 0.1617 - accuracy: 0.940 - ETA: 14s - loss: 0.1687 - accuracy: 0.936 - ETA: 13s - loss: 0.1687 - accuracy: 0.935 - ETA: 12s - loss: 0.1694 - accuracy: 0.935 - ETA: 11s - loss: 0.1701 - accuracy: 0.935 - ETA: 11s - loss: 0.1701 - accuracy: 0.935 - ETA: 11s - loss: 0.1701 - accuracy: 0.936 - ETA: 10s - loss: 0.1705 - accuracy: 0.936 - ETA: 9s - loss: 0.1705 - accuracy: 0.936 - ETA: 9s - loss: 0.1712 - accuracy: 0.93 - ETA: 8s - loss: 0.1724 - accuracy: 0.93 - ETA: 7s - loss: 0.1736 - accuracy: 0.93 - ETA: 7s - loss: 0.1748 - accuracy: 0.93 - ETA: 6s - loss: 0.1750 - accuracy: 0.93 - ETA: 5s - loss: 0.1730 - accuracy: 0.93 - ETA: 5s - loss: 0.1732 - accuracy: 0.93 - ETA: 4s - loss: 0.1736 - accuracy: 0.93 - ETA: 4s - loss: 0.1720 - accuracy: 0.93 - ETA: 3s - loss: 0.1715 - accuracy: 0.93 - ETA: 2s - loss: 0.1727 - accuracy: 0.93 - ETA: 2s - loss: 0.1710 - accuracy: 0.93 - ETA: 1s - loss: 0.1721 - accuracy: 0.93 - ETA: 0s - loss: 0.1726 - accuracy: 0.93 - ETA: 0s - loss: 0.1726 - accuracy: 0.93 - 20s 19ms/step - loss: 0.1728 - accuracy: 0.9372\n",
      "Epoch 4/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.1529 - accuracy: 0.942 - ETA: 17s - loss: 0.1558 - accuracy: 0.937 - ETA: 21s - loss: 0.1553 - accuracy: 0.942 - ETA: 19s - loss: 0.1457 - accuracy: 0.948 - ETA: 18s - loss: 0.1468 - accuracy: 0.948 - ETA: 16s - loss: 0.1415 - accuracy: 0.950 - ETA: 15s - loss: 0.1392 - accuracy: 0.951 - ETA: 14s - loss: 0.1354 - accuracy: 0.952 - ETA: 14s - loss: 0.1321 - accuracy: 0.954 - ETA: 13s - loss: 0.1336 - accuracy: 0.953 - ETA: 12s - loss: 0.1304 - accuracy: 0.955 - ETA: 11s - loss: 0.1305 - accuracy: 0.956 - ETA: 11s - loss: 0.1288 - accuracy: 0.957 - ETA: 11s - loss: 0.1274 - accuracy: 0.958 - ETA: 10s - loss: 0.1265 - accuracy: 0.958 - ETA: 9s - loss: 0.1242 - accuracy: 0.959 - ETA: 9s - loss: 0.1242 - accuracy: 0.95 - ETA: 8s - loss: 0.1240 - accuracy: 0.95 - ETA: 7s - loss: 0.1249 - accuracy: 0.95 - ETA: 7s - loss: 0.1238 - accuracy: 0.95 - ETA: 6s - loss: 0.1241 - accuracy: 0.95 - ETA: 5s - loss: 0.1268 - accuracy: 0.95 - ETA: 5s - loss: 0.1286 - accuracy: 0.95 - ETA: 4s - loss: 0.1283 - accuracy: 0.95 - ETA: 4s - loss: 0.1281 - accuracy: 0.95 - ETA: 3s - loss: 0.1298 - accuracy: 0.95 - ETA: 2s - loss: 0.1292 - accuracy: 0.95 - ETA: 2s - loss: 0.1289 - accuracy: 0.95 - ETA: 1s - loss: 0.1282 - accuracy: 0.95 - ETA: 0s - loss: 0.1271 - accuracy: 0.95 - ETA: 0s - loss: 0.1281 - accuracy: 0.95 - 19s 19ms/step - loss: 0.1290 - accuracy: 0.9569\n",
      "Epoch 5/50\n",
      "1008/1008 [==============================] - ETA: 36s - loss: 0.0943 - accuracy: 0.973 - ETA: 26s - loss: 0.0919 - accuracy: 0.973 - ETA: 22s - loss: 0.0886 - accuracy: 0.976 - ETA: 20s - loss: 0.0951 - accuracy: 0.973 - ETA: 18s - loss: 0.0936 - accuracy: 0.975 - ETA: 17s - loss: 0.0945 - accuracy: 0.974 - ETA: 16s - loss: 0.0997 - accuracy: 0.971 - ETA: 15s - loss: 0.0989 - accuracy: 0.972 - ETA: 14s - loss: 0.0999 - accuracy: 0.970 - ETA: 15s - loss: 0.1035 - accuracy: 0.969 - ETA: 14s - loss: 0.1036 - accuracy: 0.968 - ETA: 13s - loss: 0.1029 - accuracy: 0.969 - ETA: 12s - loss: 0.1026 - accuracy: 0.969 - ETA: 11s - loss: 0.1026 - accuracy: 0.969 - ETA: 11s - loss: 0.1031 - accuracy: 0.970 - ETA: 10s - loss: 0.1039 - accuracy: 0.970 - ETA: 10s - loss: 0.1032 - accuracy: 0.970 - ETA: 9s - loss: 0.1031 - accuracy: 0.969 - ETA: 9s - loss: 0.1033 - accuracy: 0.96 - ETA: 8s - loss: 0.1038 - accuracy: 0.96 - ETA: 7s - loss: 0.1033 - accuracy: 0.96 - ETA: 6s - loss: 0.1029 - accuracy: 0.96 - ETA: 6s - loss: 0.1035 - accuracy: 0.96 - ETA: 5s - loss: 0.1035 - accuracy: 0.97 - ETA: 4s - loss: 0.1039 - accuracy: 0.97 - ETA: 3s - loss: 0.1040 - accuracy: 0.97 - ETA: 3s - loss: 0.1037 - accuracy: 0.97 - ETA: 2s - loss: 0.1032 - accuracy: 0.97 - ETA: 1s - loss: 0.1041 - accuracy: 0.96 - ETA: 1s - loss: 0.1040 - accuracy: 0.96 - ETA: 0s - loss: 0.1030 - accuracy: 0.97 - 23s 23ms/step - loss: 0.1040 - accuracy: 0.9700\n",
      "Epoch 6/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 23s - loss: 0.1220 - accuracy: 0.964 - ETA: 21s - loss: 0.1191 - accuracy: 0.962 - ETA: 20s - loss: 0.1085 - accuracy: 0.970 - ETA: 23s - loss: 0.0967 - accuracy: 0.974 - ETA: 21s - loss: 0.0890 - accuracy: 0.977 - ETA: 19s - loss: 0.0862 - accuracy: 0.978 - ETA: 18s - loss: 0.0838 - accuracy: 0.979 - ETA: 17s - loss: 0.0823 - accuracy: 0.979 - ETA: 16s - loss: 0.0815 - accuracy: 0.980 - ETA: 15s - loss: 0.0796 - accuracy: 0.980 - ETA: 14s - loss: 0.0799 - accuracy: 0.980 - ETA: 13s - loss: 0.0783 - accuracy: 0.980 - ETA: 13s - loss: 0.0779 - accuracy: 0.980 - ETA: 12s - loss: 0.0769 - accuracy: 0.981 - ETA: 12s - loss: 0.0766 - accuracy: 0.981 - ETA: 11s - loss: 0.0764 - accuracy: 0.981 - ETA: 10s - loss: 0.0763 - accuracy: 0.981 - ETA: 9s - loss: 0.0757 - accuracy: 0.982 - ETA: 8s - loss: 0.0750 - accuracy: 0.98 - ETA: 8s - loss: 0.0744 - accuracy: 0.98 - ETA: 7s - loss: 0.0763 - accuracy: 0.98 - ETA: 6s - loss: 0.0763 - accuracy: 0.98 - ETA: 6s - loss: 0.0765 - accuracy: 0.98 - ETA: 5s - loss: 0.0769 - accuracy: 0.98 - ETA: 4s - loss: 0.0768 - accuracy: 0.98 - ETA: 3s - loss: 0.0766 - accuracy: 0.98 - ETA: 3s - loss: 0.0768 - accuracy: 0.98 - ETA: 2s - loss: 0.0764 - accuracy: 0.98 - ETA: 1s - loss: 0.0769 - accuracy: 0.98 - ETA: 1s - loss: 0.0764 - accuracy: 0.98 - ETA: 0s - loss: 0.0768 - accuracy: 0.98 - 22s 22ms/step - loss: 0.0766 - accuracy: 0.9821\n",
      "Epoch 7/50\n",
      "1008/1008 [==============================] - ETA: 18s - loss: 0.0448 - accuracy: 1.000 - ETA: 18s - loss: 0.0555 - accuracy: 0.991 - ETA: 17s - loss: 0.0611 - accuracy: 0.985 - ETA: 16s - loss: 0.0656 - accuracy: 0.982 - ETA: 15s - loss: 0.0649 - accuracy: 0.983 - ETA: 15s - loss: 0.0662 - accuracy: 0.982 - ETA: 14s - loss: 0.0631 - accuracy: 0.985 - ETA: 14s - loss: 0.0647 - accuracy: 0.984 - ETA: 13s - loss: 0.0624 - accuracy: 0.984 - ETA: 14s - loss: 0.0625 - accuracy: 0.985 - ETA: 14s - loss: 0.0618 - accuracy: 0.985 - ETA: 13s - loss: 0.0612 - accuracy: 0.986 - ETA: 13s - loss: 0.0616 - accuracy: 0.985 - ETA: 12s - loss: 0.0610 - accuracy: 0.986 - ETA: 11s - loss: 0.0611 - accuracy: 0.986 - ETA: 10s - loss: 0.0622 - accuracy: 0.986 - ETA: 10s - loss: 0.0623 - accuracy: 0.986 - ETA: 9s - loss: 0.0626 - accuracy: 0.986 - ETA: 8s - loss: 0.0615 - accuracy: 0.98 - ETA: 8s - loss: 0.0616 - accuracy: 0.98 - ETA: 7s - loss: 0.0610 - accuracy: 0.98 - ETA: 6s - loss: 0.0601 - accuracy: 0.98 - ETA: 6s - loss: 0.0605 - accuracy: 0.98 - ETA: 5s - loss: 0.0592 - accuracy: 0.98 - ETA: 4s - loss: 0.0585 - accuracy: 0.98 - ETA: 3s - loss: 0.0582 - accuracy: 0.98 - ETA: 3s - loss: 0.0578 - accuracy: 0.98 - ETA: 2s - loss: 0.0579 - accuracy: 0.98 - ETA: 1s - loss: 0.0579 - accuracy: 0.98 - ETA: 1s - loss: 0.0576 - accuracy: 0.98 - ETA: 0s - loss: 0.0580 - accuracy: 0.98 - 22s 22ms/step - loss: 0.0584 - accuracy: 0.9881\n",
      "Epoch 8/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0467 - accuracy: 1.000 - ETA: 16s - loss: 0.0394 - accuracy: 1.000 - ETA: 15s - loss: 0.0397 - accuracy: 0.998 - ETA: 15s - loss: 0.0360 - accuracy: 0.998 - ETA: 14s - loss: 0.0387 - accuracy: 0.997 - ETA: 16s - loss: 0.0440 - accuracy: 0.994 - ETA: 15s - loss: 0.0447 - accuracy: 0.994 - ETA: 14s - loss: 0.0430 - accuracy: 0.995 - ETA: 14s - loss: 0.0434 - accuracy: 0.996 - ETA: 13s - loss: 0.0429 - accuracy: 0.996 - ETA: 12s - loss: 0.0422 - accuracy: 0.996 - ETA: 12s - loss: 0.0420 - accuracy: 0.996 - ETA: 11s - loss: 0.0414 - accuracy: 0.996 - ETA: 10s - loss: 0.0406 - accuracy: 0.996 - ETA: 11s - loss: 0.0402 - accuracy: 0.996 - ETA: 10s - loss: 0.0410 - accuracy: 0.996 - ETA: 9s - loss: 0.0409 - accuracy: 0.996 - ETA: 9s - loss: 0.0417 - accuracy: 0.99 - ETA: 8s - loss: 0.0415 - accuracy: 0.99 - ETA: 7s - loss: 0.0421 - accuracy: 0.99 - ETA: 7s - loss: 0.0428 - accuracy: 0.99 - ETA: 6s - loss: 0.0426 - accuracy: 0.99 - ETA: 5s - loss: 0.0429 - accuracy: 0.99 - ETA: 5s - loss: 0.0435 - accuracy: 0.99 - ETA: 4s - loss: 0.0434 - accuracy: 0.99 - ETA: 3s - loss: 0.0437 - accuracy: 0.99 - ETA: 3s - loss: 0.0432 - accuracy: 0.99 - ETA: 2s - loss: 0.0434 - accuracy: 0.99 - ETA: 1s - loss: 0.0446 - accuracy: 0.99 - ETA: 1s - loss: 0.0442 - accuracy: 0.99 - ETA: 0s - loss: 0.0437 - accuracy: 0.99 - 22s 22ms/step - loss: 0.0438 - accuracy: 0.9943\n",
      "Epoch 9/50\n",
      "1008/1008 [==============================] - ETA: 24s - loss: 0.0242 - accuracy: 1.000 - ETA: 22s - loss: 0.0310 - accuracy: 0.997 - ETA: 21s - loss: 0.0326 - accuracy: 0.998 - ETA: 20s - loss: 0.0320 - accuracy: 0.998 - ETA: 18s - loss: 0.0303 - accuracy: 0.999 - ETA: 18s - loss: 0.0297 - accuracy: 0.998 - ETA: 17s - loss: 0.0285 - accuracy: 0.998 - ETA: 16s - loss: 0.0307 - accuracy: 0.998 - ETA: 16s - loss: 0.0293 - accuracy: 0.998 - ETA: 16s - loss: 0.0299 - accuracy: 0.997 - ETA: 15s - loss: 0.0309 - accuracy: 0.997 - ETA: 14s - loss: 0.0314 - accuracy: 0.997 - ETA: 13s - loss: 0.0301 - accuracy: 0.997 - ETA: 12s - loss: 0.0295 - accuracy: 0.997 - ETA: 11s - loss: 0.0293 - accuracy: 0.997 - ETA: 11s - loss: 0.0290 - accuracy: 0.998 - ETA: 10s - loss: 0.0289 - accuracy: 0.998 - ETA: 9s - loss: 0.0293 - accuracy: 0.998 - ETA: 9s - loss: 0.0287 - accuracy: 0.99 - ETA: 8s - loss: 0.0284 - accuracy: 0.99 - ETA: 7s - loss: 0.0280 - accuracy: 0.99 - ETA: 6s - loss: 0.0278 - accuracy: 0.99 - ETA: 6s - loss: 0.0277 - accuracy: 0.99 - ETA: 5s - loss: 0.0280 - accuracy: 0.99 - ETA: 4s - loss: 0.0283 - accuracy: 0.99 - ETA: 3s - loss: 0.0281 - accuracy: 0.99 - ETA: 3s - loss: 0.0284 - accuracy: 0.99 - ETA: 2s - loss: 0.0281 - accuracy: 0.99 - ETA: 1s - loss: 0.0286 - accuracy: 0.99 - ETA: 1s - loss: 0.0286 - accuracy: 0.99 - ETA: 0s - loss: 0.0285 - accuracy: 0.99 - 22s 22ms/step - loss: 0.0284 - accuracy: 0.9986\n",
      "Epoch 10/50\n",
      "1008/1008 [==============================] - ETA: 18s - loss: 0.0254 - accuracy: 1.000 - ETA: 17s - loss: 0.0256 - accuracy: 1.000 - ETA: 16s - loss: 0.0251 - accuracy: 1.000 - ETA: 16s - loss: 0.0291 - accuracy: 0.997 - ETA: 15s - loss: 0.0275 - accuracy: 0.997 - ETA: 14s - loss: 0.0260 - accuracy: 0.997 - ETA: 16s - loss: 0.0246 - accuracy: 0.998 - ETA: 15s - loss: 0.0246 - accuracy: 0.998 - ETA: 14s - loss: 0.0242 - accuracy: 0.998 - ETA: 14s - loss: 0.0252 - accuracy: 0.997 - ETA: 13s - loss: 0.0247 - accuracy: 0.998 - ETA: 12s - loss: 0.0240 - accuracy: 0.998 - ETA: 11s - loss: 0.0233 - accuracy: 0.998 - ETA: 10s - loss: 0.0230 - accuracy: 0.998 - ETA: 10s - loss: 0.0232 - accuracy: 0.998 - ETA: 9s - loss: 0.0235 - accuracy: 0.998 - ETA: 9s - loss: 0.0230 - accuracy: 0.99 - ETA: 8s - loss: 0.0231 - accuracy: 0.99 - ETA: 8s - loss: 0.0235 - accuracy: 0.99 - ETA: 7s - loss: 0.0237 - accuracy: 0.99 - ETA: 6s - loss: 0.0235 - accuracy: 0.99 - ETA: 6s - loss: 0.0231 - accuracy: 0.99 - ETA: 5s - loss: 0.0229 - accuracy: 0.99 - ETA: 4s - loss: 0.0226 - accuracy: 0.99 - ETA: 4s - loss: 0.0223 - accuracy: 0.99 - ETA: 3s - loss: 0.0222 - accuracy: 0.99 - ETA: 2s - loss: 0.0221 - accuracy: 0.99 - ETA: 2s - loss: 0.0220 - accuracy: 0.99 - ETA: 1s - loss: 0.0217 - accuracy: 0.99 - ETA: 0s - loss: 0.0215 - accuracy: 0.99 - ETA: 0s - loss: 0.0215 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0216 - accuracy: 0.9991\n",
      "Epoch 11/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0122 - accuracy: 1.000 - ETA: 16s - loss: 0.0213 - accuracy: 0.995 - ETA: 16s - loss: 0.0190 - accuracy: 0.997 - ETA: 18s - loss: 0.0180 - accuracy: 0.997 - ETA: 18s - loss: 0.0187 - accuracy: 0.998 - ETA: 17s - loss: 0.0175 - accuracy: 0.998 - ETA: 16s - loss: 0.0165 - accuracy: 0.998 - ETA: 15s - loss: 0.0175 - accuracy: 0.998 - ETA: 15s - loss: 0.0175 - accuracy: 0.999 - ETA: 14s - loss: 0.0172 - accuracy: 0.999 - ETA: 13s - loss: 0.0174 - accuracy: 0.999 - ETA: 12s - loss: 0.0179 - accuracy: 0.999 - ETA: 12s - loss: 0.0175 - accuracy: 0.999 - ETA: 12s - loss: 0.0176 - accuracy: 0.999 - ETA: 11s - loss: 0.0172 - accuracy: 0.999 - ETA: 10s - loss: 0.0170 - accuracy: 0.999 - ETA: 9s - loss: 0.0174 - accuracy: 0.998 - ETA: 9s - loss: 0.0171 - accuracy: 0.99 - ETA: 8s - loss: 0.0174 - accuracy: 0.99 - ETA: 7s - loss: 0.0175 - accuracy: 0.99 - ETA: 7s - loss: 0.0178 - accuracy: 0.99 - ETA: 6s - loss: 0.0179 - accuracy: 0.99 - ETA: 5s - loss: 0.0178 - accuracy: 0.99 - ETA: 5s - loss: 0.0176 - accuracy: 0.99 - ETA: 4s - loss: 0.0173 - accuracy: 0.99 - ETA: 3s - loss: 0.0174 - accuracy: 0.99 - ETA: 3s - loss: 0.0173 - accuracy: 0.99 - ETA: 2s - loss: 0.0175 - accuracy: 0.99 - ETA: 1s - loss: 0.0175 - accuracy: 0.99 - ETA: 1s - loss: 0.0174 - accuracy: 0.99 - ETA: 0s - loss: 0.0175 - accuracy: 0.99 - 21s 21ms/step - loss: 0.0175 - accuracy: 0.9990\n",
      "Epoch 12/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0098 - accuracy: 1.000 - ETA: 26s - loss: 0.0098 - accuracy: 1.000 - ETA: 23s - loss: 0.0098 - accuracy: 1.000 - ETA: 21s - loss: 0.0127 - accuracy: 0.998 - ETA: 19s - loss: 0.0120 - accuracy: 0.999 - ETA: 18s - loss: 0.0123 - accuracy: 0.999 - ETA: 16s - loss: 0.0124 - accuracy: 0.999 - ETA: 15s - loss: 0.0118 - accuracy: 0.999 - ETA: 14s - loss: 0.0119 - accuracy: 0.999 - ETA: 14s - loss: 0.0126 - accuracy: 0.999 - ETA: 14s - loss: 0.0129 - accuracy: 0.999 - ETA: 13s - loss: 0.0130 - accuracy: 0.999 - ETA: 12s - loss: 0.0130 - accuracy: 0.999 - ETA: 11s - loss: 0.0127 - accuracy: 0.999 - ETA: 11s - loss: 0.0134 - accuracy: 0.998 - ETA: 10s - loss: 0.0141 - accuracy: 0.998 - ETA: 9s - loss: 0.0139 - accuracy: 0.998 - ETA: 8s - loss: 0.0138 - accuracy: 0.99 - ETA: 8s - loss: 0.0137 - accuracy: 0.99 - ETA: 7s - loss: 0.0137 - accuracy: 0.99 - ETA: 7s - loss: 0.0134 - accuracy: 0.99 - ETA: 6s - loss: 0.0138 - accuracy: 0.99 - ETA: 5s - loss: 0.0138 - accuracy: 0.99 - ETA: 5s - loss: 0.0138 - accuracy: 0.99 - ETA: 4s - loss: 0.0141 - accuracy: 0.99 - ETA: 3s - loss: 0.0142 - accuracy: 0.99 - ETA: 2s - loss: 0.0141 - accuracy: 0.99 - ETA: 2s - loss: 0.0142 - accuracy: 0.99 - ETA: 1s - loss: 0.0140 - accuracy: 0.99 - ETA: 0s - loss: 0.0141 - accuracy: 0.99 - ETA: 0s - loss: 0.0140 - accuracy: 0.99 - 21s 21ms/step - loss: 0.0141 - accuracy: 0.9991\n",
      "Epoch 13/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0102 - accuracy: 1.000 - ETA: 16s - loss: 0.0122 - accuracy: 1.000 - ETA: 15s - loss: 0.0112 - accuracy: 1.000 - ETA: 15s - loss: 0.0130 - accuracy: 1.000 - ETA: 14s - loss: 0.0123 - accuracy: 1.000 - ETA: 14s - loss: 0.0122 - accuracy: 1.000 - ETA: 13s - loss: 0.0130 - accuracy: 1.000 - ETA: 13s - loss: 0.0122 - accuracy: 1.000 - ETA: 12s - loss: 0.0118 - accuracy: 1.000 - ETA: 12s - loss: 0.0125 - accuracy: 1.000 - ETA: 11s - loss: 0.0122 - accuracy: 1.000 - ETA: 11s - loss: 0.0122 - accuracy: 1.000 - ETA: 11s - loss: 0.0126 - accuracy: 1.000 - ETA: 10s - loss: 0.0123 - accuracy: 1.000 - ETA: 10s - loss: 0.0123 - accuracy: 1.000 - ETA: 9s - loss: 0.0121 - accuracy: 1.000 - ETA: 8s - loss: 0.0121 - accuracy: 1.00 - ETA: 8s - loss: 0.0120 - accuracy: 1.00 - ETA: 7s - loss: 0.0124 - accuracy: 0.99 - ETA: 6s - loss: 0.0122 - accuracy: 0.99 - ETA: 6s - loss: 0.0121 - accuracy: 0.99 - ETA: 5s - loss: 0.0120 - accuracy: 0.99 - ETA: 5s - loss: 0.0125 - accuracy: 0.99 - ETA: 4s - loss: 0.0125 - accuracy: 0.99 - ETA: 4s - loss: 0.0123 - accuracy: 0.99 - ETA: 3s - loss: 0.0128 - accuracy: 0.99 - ETA: 2s - loss: 0.0127 - accuracy: 0.99 - ETA: 2s - loss: 0.0127 - accuracy: 0.99 - ETA: 1s - loss: 0.0127 - accuracy: 0.99 - ETA: 0s - loss: 0.0128 - accuracy: 0.99 - ETA: 0s - loss: 0.0126 - accuracy: 0.99 - 19s 19ms/step - loss: 0.0126 - accuracy: 0.9994\n",
      "Epoch 14/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0089 - accuracy: 1.000 - ETA: 16s - loss: 0.0085 - accuracy: 1.000 - ETA: 21s - loss: 0.0080 - accuracy: 1.000 - ETA: 19s - loss: 0.0086 - accuracy: 1.000 - ETA: 18s - loss: 0.0109 - accuracy: 0.997 - ETA: 17s - loss: 0.0103 - accuracy: 0.997 - ETA: 16s - loss: 0.0105 - accuracy: 0.998 - ETA: 15s - loss: 0.0103 - accuracy: 0.998 - ETA: 14s - loss: 0.0103 - accuracy: 0.998 - ETA: 13s - loss: 0.0101 - accuracy: 0.998 - ETA: 12s - loss: 0.0100 - accuracy: 0.998 - ETA: 11s - loss: 0.0095 - accuracy: 0.998 - ETA: 11s - loss: 0.0096 - accuracy: 0.999 - ETA: 10s - loss: 0.0094 - accuracy: 0.999 - ETA: 9s - loss: 0.0091 - accuracy: 0.999 - ETA: 9s - loss: 0.0094 - accuracy: 0.99 - ETA: 8s - loss: 0.0094 - accuracy: 0.99 - ETA: 8s - loss: 0.0093 - accuracy: 0.99 - ETA: 7s - loss: 0.0092 - accuracy: 0.99 - ETA: 7s - loss: 0.0093 - accuracy: 0.99 - ETA: 6s - loss: 0.0091 - accuracy: 0.99 - ETA: 5s - loss: 0.0089 - accuracy: 0.99 - ETA: 5s - loss: 0.0089 - accuracy: 0.99 - ETA: 4s - loss: 0.0089 - accuracy: 0.99 - ETA: 3s - loss: 0.0088 - accuracy: 0.99 - ETA: 3s - loss: 0.0089 - accuracy: 0.99 - ETA: 2s - loss: 0.0089 - accuracy: 0.99 - ETA: 2s - loss: 0.0090 - accuracy: 0.99 - ETA: 1s - loss: 0.0090 - accuracy: 0.99 - ETA: 0s - loss: 0.0089 - accuracy: 0.99 - ETA: 0s - loss: 0.0089 - accuracy: 0.99 - 19s 19ms/step - loss: 0.0089 - accuracy: 0.9996\n",
      "Epoch 15/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0089 - accuracy: 1.000 - ETA: 16s - loss: 0.0095 - accuracy: 1.000 - ETA: 16s - loss: 0.0092 - accuracy: 1.000 - ETA: 19s - loss: 0.0088 - accuracy: 1.000 - ETA: 18s - loss: 0.0082 - accuracy: 1.000 - ETA: 17s - loss: 0.0088 - accuracy: 1.000 - ETA: 16s - loss: 0.0084 - accuracy: 1.000 - ETA: 15s - loss: 0.0085 - accuracy: 1.000 - ETA: 14s - loss: 0.0087 - accuracy: 1.000 - ETA: 13s - loss: 0.0085 - accuracy: 1.000 - ETA: 12s - loss: 0.0093 - accuracy: 1.000 - ETA: 12s - loss: 0.0094 - accuracy: 1.000 - ETA: 11s - loss: 0.0091 - accuracy: 1.000 - ETA: 10s - loss: 0.0089 - accuracy: 1.000 - ETA: 9s - loss: 0.0088 - accuracy: 1.000 - ETA: 9s - loss: 0.0088 - accuracy: 1.00 - ETA: 8s - loss: 0.0087 - accuracy: 1.00 - ETA: 8s - loss: 0.0087 - accuracy: 1.00 - ETA: 7s - loss: 0.0089 - accuracy: 1.00 - ETA: 7s - loss: 0.0089 - accuracy: 1.00 - ETA: 6s - loss: 0.0091 - accuracy: 1.00 - ETA: 6s - loss: 0.0091 - accuracy: 1.00 - ETA: 5s - loss: 0.0090 - accuracy: 1.00 - ETA: 4s - loss: 0.0091 - accuracy: 1.00 - ETA: 4s - loss: 0.0090 - accuracy: 1.00 - ETA: 3s - loss: 0.0091 - accuracy: 1.00 - ETA: 2s - loss: 0.0089 - accuracy: 1.00 - ETA: 2s - loss: 0.0088 - accuracy: 1.00 - ETA: 1s - loss: 0.0089 - accuracy: 1.00 - ETA: 0s - loss: 0.0089 - accuracy: 1.00 - ETA: 0s - loss: 0.0089 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0089 - accuracy: 1.0000\n",
      "Epoch 16/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 18s - loss: 0.0068 - accuracy: 1.000 - ETA: 17s - loss: 0.0072 - accuracy: 1.000 - ETA: 16s - loss: 0.0059 - accuracy: 1.000 - ETA: 15s - loss: 0.0059 - accuracy: 1.000 - ETA: 15s - loss: 0.0065 - accuracy: 1.000 - ETA: 14s - loss: 0.0071 - accuracy: 1.000 - ETA: 13s - loss: 0.0073 - accuracy: 1.000 - ETA: 13s - loss: 0.0072 - accuracy: 1.000 - ETA: 12s - loss: 0.0070 - accuracy: 1.000 - ETA: 12s - loss: 0.0071 - accuracy: 1.000 - ETA: 12s - loss: 0.0075 - accuracy: 1.000 - ETA: 12s - loss: 0.0075 - accuracy: 1.000 - ETA: 11s - loss: 0.0079 - accuracy: 1.000 - ETA: 10s - loss: 0.0078 - accuracy: 1.000 - ETA: 10s - loss: 0.0076 - accuracy: 1.000 - ETA: 9s - loss: 0.0074 - accuracy: 1.000 - ETA: 8s - loss: 0.0073 - accuracy: 1.00 - ETA: 8s - loss: 0.0072 - accuracy: 1.00 - ETA: 7s - loss: 0.0073 - accuracy: 1.00 - ETA: 7s - loss: 0.0073 - accuracy: 1.00 - ETA: 6s - loss: 0.0072 - accuracy: 1.00 - ETA: 6s - loss: 0.0070 - accuracy: 1.00 - ETA: 5s - loss: 0.0070 - accuracy: 1.00 - ETA: 4s - loss: 0.0071 - accuracy: 1.00 - ETA: 4s - loss: 0.0071 - accuracy: 1.00 - ETA: 3s - loss: 0.0071 - accuracy: 1.00 - ETA: 2s - loss: 0.0072 - accuracy: 1.00 - ETA: 2s - loss: 0.0073 - accuracy: 1.00 - ETA: 1s - loss: 0.0073 - accuracy: 1.00 - ETA: 0s - loss: 0.0073 - accuracy: 1.00 - ETA: 0s - loss: 0.0072 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0072 - accuracy: 1.0000\n",
      "Epoch 17/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0035 - accuracy: 1.000 - ETA: 17s - loss: 0.0036 - accuracy: 1.000 - ETA: 17s - loss: 0.0042 - accuracy: 1.000 - ETA: 16s - loss: 0.0040 - accuracy: 1.000 - ETA: 16s - loss: 0.0056 - accuracy: 1.000 - ETA: 15s - loss: 0.0055 - accuracy: 1.000 - ETA: 15s - loss: 0.0054 - accuracy: 1.000 - ETA: 14s - loss: 0.0054 - accuracy: 1.000 - ETA: 14s - loss: 0.0052 - accuracy: 1.000 - ETA: 14s - loss: 0.0060 - accuracy: 1.000 - ETA: 13s - loss: 0.0064 - accuracy: 1.000 - ETA: 13s - loss: 0.0063 - accuracy: 1.000 - ETA: 12s - loss: 0.0060 - accuracy: 1.000 - ETA: 11s - loss: 0.0062 - accuracy: 1.000 - ETA: 10s - loss: 0.0060 - accuracy: 1.000 - ETA: 10s - loss: 0.0062 - accuracy: 1.000 - ETA: 9s - loss: 0.0062 - accuracy: 1.000 - ETA: 8s - loss: 0.0061 - accuracy: 1.00 - ETA: 7s - loss: 0.0062 - accuracy: 1.00 - ETA: 7s - loss: 0.0061 - accuracy: 1.00 - ETA: 6s - loss: 0.0063 - accuracy: 1.00 - ETA: 6s - loss: 0.0063 - accuracy: 1.00 - ETA: 5s - loss: 0.0068 - accuracy: 1.00 - ETA: 4s - loss: 0.0070 - accuracy: 1.00 - ETA: 4s - loss: 0.0069 - accuracy: 1.00 - ETA: 3s - loss: 0.0070 - accuracy: 1.00 - ETA: 2s - loss: 0.0071 - accuracy: 1.00 - ETA: 2s - loss: 0.0071 - accuracy: 1.00 - ETA: 1s - loss: 0.0073 - accuracy: 1.00 - ETA: 0s - loss: 0.0072 - accuracy: 1.00 - ETA: 0s - loss: 0.0071 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0071 - accuracy: 1.0000\n",
      "Epoch 18/50\n",
      "1008/1008 [==============================] - ETA: 18s - loss: 0.0037 - accuracy: 1.000 - ETA: 17s - loss: 0.0057 - accuracy: 1.000 - ETA: 17s - loss: 0.0052 - accuracy: 1.000 - ETA: 16s - loss: 0.0051 - accuracy: 1.000 - ETA: 15s - loss: 0.0064 - accuracy: 1.000 - ETA: 15s - loss: 0.0075 - accuracy: 1.000 - ETA: 14s - loss: 0.0079 - accuracy: 1.000 - ETA: 13s - loss: 0.0075 - accuracy: 1.000 - ETA: 14s - loss: 0.0072 - accuracy: 1.000 - ETA: 14s - loss: 0.0070 - accuracy: 1.000 - ETA: 13s - loss: 0.0068 - accuracy: 1.000 - ETA: 12s - loss: 0.0066 - accuracy: 1.000 - ETA: 11s - loss: 0.0063 - accuracy: 1.000 - ETA: 11s - loss: 0.0061 - accuracy: 1.000 - ETA: 10s - loss: 0.0062 - accuracy: 1.000 - ETA: 9s - loss: 0.0061 - accuracy: 1.000 - ETA: 9s - loss: 0.0061 - accuracy: 1.00 - ETA: 8s - loss: 0.0060 - accuracy: 1.00 - ETA: 8s - loss: 0.0060 - accuracy: 1.00 - ETA: 7s - loss: 0.0061 - accuracy: 1.00 - ETA: 6s - loss: 0.0064 - accuracy: 1.00 - ETA: 6s - loss: 0.0064 - accuracy: 1.00 - ETA: 5s - loss: 0.0064 - accuracy: 1.00 - ETA: 4s - loss: 0.0063 - accuracy: 1.00 - ETA: 4s - loss: 0.0064 - accuracy: 1.00 - ETA: 3s - loss: 0.0064 - accuracy: 1.00 - ETA: 2s - loss: 0.0063 - accuracy: 1.00 - ETA: 2s - loss: 0.0062 - accuracy: 1.00 - ETA: 1s - loss: 0.0062 - accuracy: 1.00 - ETA: 0s - loss: 0.0061 - accuracy: 1.00 - ETA: 0s - loss: 0.0062 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0062 - accuracy: 1.0000\n",
      "Epoch 19/50\n",
      "1008/1008 [==============================] - ETA: 26s - loss: 0.0100 - accuracy: 1.000 - ETA: 21s - loss: 0.0075 - accuracy: 1.000 - ETA: 19s - loss: 0.0062 - accuracy: 1.000 - ETA: 17s - loss: 0.0060 - accuracy: 1.000 - ETA: 19s - loss: 0.0066 - accuracy: 1.000 - ETA: 18s - loss: 0.0073 - accuracy: 1.000 - ETA: 17s - loss: 0.0071 - accuracy: 1.000 - ETA: 16s - loss: 0.0070 - accuracy: 1.000 - ETA: 15s - loss: 0.0065 - accuracy: 1.000 - ETA: 14s - loss: 0.0067 - accuracy: 1.000 - ETA: 13s - loss: 0.0066 - accuracy: 1.000 - ETA: 12s - loss: 0.0064 - accuracy: 1.000 - ETA: 11s - loss: 0.0066 - accuracy: 1.000 - ETA: 11s - loss: 0.0065 - accuracy: 1.000 - ETA: 10s - loss: 0.0063 - accuracy: 1.000 - ETA: 9s - loss: 0.0062 - accuracy: 1.000 - ETA: 8s - loss: 0.0062 - accuracy: 1.00 - ETA: 8s - loss: 0.0062 - accuracy: 1.00 - ETA: 7s - loss: 0.0063 - accuracy: 1.00 - ETA: 7s - loss: 0.0062 - accuracy: 1.00 - ETA: 6s - loss: 0.0061 - accuracy: 1.00 - ETA: 5s - loss: 0.0061 - accuracy: 1.00 - ETA: 5s - loss: 0.0061 - accuracy: 1.00 - ETA: 4s - loss: 0.0060 - accuracy: 1.00 - ETA: 4s - loss: 0.0060 - accuracy: 1.00 - ETA: 3s - loss: 0.0058 - accuracy: 1.00 - ETA: 2s - loss: 0.0060 - accuracy: 1.00 - ETA: 2s - loss: 0.0059 - accuracy: 1.00 - ETA: 1s - loss: 0.0059 - accuracy: 1.00 - ETA: 0s - loss: 0.0059 - accuracy: 1.00 - ETA: 0s - loss: 0.0058 - accuracy: 1.00 - 19s 19ms/step - loss: 0.0057 - accuracy: 1.0000\n",
      "Epoch 20/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0048 - accuracy: 1.000 - ETA: 22s - loss: 0.0048 - accuracy: 1.000 - ETA: 20s - loss: 0.0068 - accuracy: 1.000 - ETA: 18s - loss: 0.0062 - accuracy: 1.000 - ETA: 17s - loss: 0.0057 - accuracy: 1.000 - ETA: 16s - loss: 0.0064 - accuracy: 1.000 - ETA: 15s - loss: 0.0062 - accuracy: 1.000 - ETA: 14s - loss: 0.0061 - accuracy: 1.000 - ETA: 13s - loss: 0.0058 - accuracy: 1.000 - ETA: 13s - loss: 0.0059 - accuracy: 1.000 - ETA: 12s - loss: 0.0056 - accuracy: 1.000 - ETA: 12s - loss: 0.0055 - accuracy: 1.000 - ETA: 11s - loss: 0.0055 - accuracy: 1.000 - ETA: 11s - loss: 0.0055 - accuracy: 1.000 - ETA: 10s - loss: 0.0055 - accuracy: 1.000 - ETA: 9s - loss: 0.0054 - accuracy: 1.000 - ETA: 9s - loss: 0.0055 - accuracy: 1.00 - ETA: 8s - loss: 0.0054 - accuracy: 1.00 - ETA: 7s - loss: 0.0053 - accuracy: 1.00 - ETA: 7s - loss: 0.0052 - accuracy: 1.00 - ETA: 6s - loss: 0.0051 - accuracy: 1.00 - ETA: 5s - loss: 0.0050 - accuracy: 1.00 - ETA: 5s - loss: 0.0053 - accuracy: 0.99 - ETA: 4s - loss: 0.0053 - accuracy: 0.99 - ETA: 4s - loss: 0.0052 - accuracy: 0.99 - ETA: 3s - loss: 0.0051 - accuracy: 0.99 - ETA: 2s - loss: 0.0051 - accuracy: 0.99 - ETA: 2s - loss: 0.0050 - accuracy: 0.99 - ETA: 1s - loss: 0.0053 - accuracy: 0.99 - ETA: 0s - loss: 0.0053 - accuracy: 0.99 - ETA: 0s - loss: 0.0052 - accuracy: 0.99 - 19s 19ms/step - loss: 0.0052 - accuracy: 0.9999\n",
      "Epoch 21/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 19s - loss: 0.0030 - accuracy: 1.000 - ETA: 24s - loss: 0.0028 - accuracy: 1.000 - ETA: 21s - loss: 0.0032 - accuracy: 1.000 - ETA: 19s - loss: 0.0040 - accuracy: 1.000 - ETA: 17s - loss: 0.0042 - accuracy: 1.000 - ETA: 16s - loss: 0.0039 - accuracy: 1.000 - ETA: 15s - loss: 0.0041 - accuracy: 1.000 - ETA: 14s - loss: 0.0040 - accuracy: 1.000 - ETA: 13s - loss: 0.0038 - accuracy: 1.000 - ETA: 12s - loss: 0.0038 - accuracy: 1.000 - ETA: 12s - loss: 0.0039 - accuracy: 1.000 - ETA: 12s - loss: 0.0040 - accuracy: 1.000 - ETA: 11s - loss: 0.0041 - accuracy: 1.000 - ETA: 11s - loss: 0.0041 - accuracy: 1.000 - ETA: 10s - loss: 0.0043 - accuracy: 1.000 - ETA: 9s - loss: 0.0042 - accuracy: 1.000 - ETA: 9s - loss: 0.0043 - accuracy: 1.00 - ETA: 8s - loss: 0.0045 - accuracy: 1.00 - ETA: 7s - loss: 0.0044 - accuracy: 1.00 - ETA: 7s - loss: 0.0045 - accuracy: 1.00 - ETA: 6s - loss: 0.0047 - accuracy: 1.00 - ETA: 6s - loss: 0.0048 - accuracy: 1.00 - ETA: 5s - loss: 0.0051 - accuracy: 1.00 - ETA: 4s - loss: 0.0051 - accuracy: 1.00 - ETA: 4s - loss: 0.0052 - accuracy: 1.00 - ETA: 3s - loss: 0.0051 - accuracy: 1.00 - ETA: 2s - loss: 0.0051 - accuracy: 1.00 - ETA: 2s - loss: 0.0050 - accuracy: 1.00 - ETA: 1s - loss: 0.0051 - accuracy: 1.00 - ETA: 0s - loss: 0.0051 - accuracy: 1.00 - ETA: 0s - loss: 0.0051 - accuracy: 1.00 - 20s 19ms/step - loss: 0.0051 - accuracy: 1.0000\n",
      "Epoch 22/50\n",
      "1008/1008 [==============================] - ETA: 28s - loss: 0.0027 - accuracy: 1.000 - ETA: 26s - loss: 0.0040 - accuracy: 1.000 - ETA: 23s - loss: 0.0037 - accuracy: 1.000 - ETA: 21s - loss: 0.0048 - accuracy: 1.000 - ETA: 19s - loss: 0.0054 - accuracy: 1.000 - ETA: 18s - loss: 0.0049 - accuracy: 1.000 - ETA: 16s - loss: 0.0047 - accuracy: 1.000 - ETA: 15s - loss: 0.0049 - accuracy: 1.000 - ETA: 14s - loss: 0.0051 - accuracy: 1.000 - ETA: 15s - loss: 0.0050 - accuracy: 1.000 - ETA: 14s - loss: 0.0048 - accuracy: 1.000 - ETA: 13s - loss: 0.0047 - accuracy: 1.000 - ETA: 12s - loss: 0.0046 - accuracy: 1.000 - ETA: 11s - loss: 0.0045 - accuracy: 1.000 - ETA: 10s - loss: 0.0049 - accuracy: 1.000 - ETA: 10s - loss: 0.0053 - accuracy: 1.000 - ETA: 9s - loss: 0.0052 - accuracy: 1.000 - ETA: 8s - loss: 0.0053 - accuracy: 1.00 - ETA: 8s - loss: 0.0052 - accuracy: 1.00 - ETA: 7s - loss: 0.0051 - accuracy: 1.00 - ETA: 7s - loss: 0.0051 - accuracy: 1.00 - ETA: 6s - loss: 0.0051 - accuracy: 1.00 - ETA: 5s - loss: 0.0050 - accuracy: 1.00 - ETA: 4s - loss: 0.0050 - accuracy: 1.00 - ETA: 4s - loss: 0.0050 - accuracy: 1.00 - ETA: 3s - loss: 0.0049 - accuracy: 1.00 - ETA: 2s - loss: 0.0049 - accuracy: 1.00 - ETA: 2s - loss: 0.0048 - accuracy: 1.00 - ETA: 1s - loss: 0.0048 - accuracy: 1.00 - ETA: 0s - loss: 0.0048 - accuracy: 1.00 - ETA: 0s - loss: 0.0048 - accuracy: 1.00 - 21s 21ms/step - loss: 0.0048 - accuracy: 1.0000\n",
      "Epoch 23/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0025 - accuracy: 1.000 - ETA: 17s - loss: 0.0026 - accuracy: 1.000 - ETA: 16s - loss: 0.0036 - accuracy: 1.000 - ETA: 15s - loss: 0.0034 - accuracy: 1.000 - ETA: 15s - loss: 0.0039 - accuracy: 1.000 - ETA: 14s - loss: 0.0037 - accuracy: 1.000 - ETA: 13s - loss: 0.0039 - accuracy: 1.000 - ETA: 15s - loss: 0.0039 - accuracy: 1.000 - ETA: 14s - loss: 0.0039 - accuracy: 1.000 - ETA: 13s - loss: 0.0046 - accuracy: 1.000 - ETA: 13s - loss: 0.0044 - accuracy: 1.000 - ETA: 12s - loss: 0.0042 - accuracy: 1.000 - ETA: 11s - loss: 0.0041 - accuracy: 1.000 - ETA: 10s - loss: 0.0040 - accuracy: 1.000 - ETA: 10s - loss: 0.0039 - accuracy: 1.000 - ETA: 10s - loss: 0.0040 - accuracy: 1.000 - ETA: 9s - loss: 0.0039 - accuracy: 1.000 - ETA: 8s - loss: 0.0040 - accuracy: 1.00 - ETA: 8s - loss: 0.0039 - accuracy: 1.00 - ETA: 7s - loss: 0.0038 - accuracy: 1.00 - ETA: 6s - loss: 0.0038 - accuracy: 1.00 - ETA: 6s - loss: 0.0037 - accuracy: 1.00 - ETA: 5s - loss: 0.0038 - accuracy: 1.00 - ETA: 4s - loss: 0.0042 - accuracy: 0.99 - ETA: 4s - loss: 0.0042 - accuracy: 0.99 - ETA: 3s - loss: 0.0042 - accuracy: 0.99 - ETA: 2s - loss: 0.0044 - accuracy: 0.99 - ETA: 2s - loss: 0.0043 - accuracy: 0.99 - ETA: 1s - loss: 0.0043 - accuracy: 0.99 - ETA: 0s - loss: 0.0043 - accuracy: 0.99 - ETA: 0s - loss: 0.0043 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0044 - accuracy: 0.9997\n",
      "Epoch 24/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0023 - accuracy: 1.000 - ETA: 17s - loss: 0.0033 - accuracy: 1.000 - ETA: 21s - loss: 0.0036 - accuracy: 1.000 - ETA: 19s - loss: 0.0038 - accuracy: 1.000 - ETA: 18s - loss: 0.0038 - accuracy: 1.000 - ETA: 17s - loss: 0.0039 - accuracy: 1.000 - ETA: 15s - loss: 0.0042 - accuracy: 1.000 - ETA: 15s - loss: 0.0040 - accuracy: 1.000 - ETA: 14s - loss: 0.0040 - accuracy: 1.000 - ETA: 13s - loss: 0.0044 - accuracy: 1.000 - ETA: 12s - loss: 0.0044 - accuracy: 1.000 - ETA: 12s - loss: 0.0043 - accuracy: 1.000 - ETA: 12s - loss: 0.0055 - accuracy: 0.999 - ETA: 11s - loss: 0.0057 - accuracy: 0.999 - ETA: 10s - loss: 0.0055 - accuracy: 0.999 - ETA: 9s - loss: 0.0055 - accuracy: 0.999 - ETA: 9s - loss: 0.0054 - accuracy: 0.99 - ETA: 8s - loss: 0.0055 - accuracy: 0.99 - ETA: 7s - loss: 0.0054 - accuracy: 0.99 - ETA: 7s - loss: 0.0054 - accuracy: 0.99 - ETA: 6s - loss: 0.0054 - accuracy: 0.99 - ETA: 6s - loss: 0.0055 - accuracy: 0.99 - ETA: 5s - loss: 0.0056 - accuracy: 0.99 - ETA: 4s - loss: 0.0055 - accuracy: 0.99 - ETA: 4s - loss: 0.0055 - accuracy: 0.99 - ETA: 3s - loss: 0.0055 - accuracy: 0.99 - ETA: 2s - loss: 0.0055 - accuracy: 0.99 - ETA: 2s - loss: 0.0055 - accuracy: 0.99 - ETA: 1s - loss: 0.0055 - accuracy: 0.99 - ETA: 0s - loss: 0.0059 - accuracy: 0.99 - ETA: 0s - loss: 0.0058 - accuracy: 0.99 - 20s 19ms/step - loss: 0.0058 - accuracy: 0.9997\n",
      "Epoch 25/50\n",
      "1008/1008 [==============================] - ETA: 39s - loss: 0.0062 - accuracy: 1.000 - ETA: 27s - loss: 0.0064 - accuracy: 1.000 - ETA: 23s - loss: 0.0058 - accuracy: 1.000 - ETA: 20s - loss: 0.0068 - accuracy: 1.000 - ETA: 18s - loss: 0.0062 - accuracy: 1.000 - ETA: 17s - loss: 0.0073 - accuracy: 1.000 - ETA: 16s - loss: 0.0065 - accuracy: 1.000 - ETA: 15s - loss: 0.0062 - accuracy: 1.000 - ETA: 14s - loss: 0.0057 - accuracy: 1.000 - ETA: 14s - loss: 0.0062 - accuracy: 1.000 - ETA: 14s - loss: 0.0066 - accuracy: 1.000 - ETA: 13s - loss: 0.0066 - accuracy: 1.000 - ETA: 12s - loss: 0.0063 - accuracy: 1.000 - ETA: 11s - loss: 0.0060 - accuracy: 1.000 - ETA: 10s - loss: 0.0057 - accuracy: 1.000 - ETA: 9s - loss: 0.0061 - accuracy: 1.000 - ETA: 9s - loss: 0.0061 - accuracy: 1.00 - ETA: 8s - loss: 0.0060 - accuracy: 1.00 - ETA: 7s - loss: 0.0058 - accuracy: 1.00 - ETA: 7s - loss: 0.0058 - accuracy: 1.00 - ETA: 6s - loss: 0.0059 - accuracy: 1.00 - ETA: 6s - loss: 0.0058 - accuracy: 1.00 - ETA: 5s - loss: 0.0058 - accuracy: 1.00 - ETA: 4s - loss: 0.0058 - accuracy: 1.00 - ETA: 4s - loss: 0.0058 - accuracy: 1.00 - ETA: 3s - loss: 0.0061 - accuracy: 0.99 - ETA: 2s - loss: 0.0061 - accuracy: 0.99 - ETA: 2s - loss: 0.0061 - accuracy: 0.99 - ETA: 1s - loss: 0.0060 - accuracy: 0.99 - ETA: 0s - loss: 0.0059 - accuracy: 0.99 - ETA: 0s - loss: 0.0058 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0058 - accuracy: 0.9997\n",
      "Epoch 26/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0086 - accuracy: 1.000 - ETA: 15s - loss: 0.0061 - accuracy: 1.000 - ETA: 15s - loss: 0.0073 - accuracy: 1.000 - ETA: 14s - loss: 0.0069 - accuracy: 1.000 - ETA: 14s - loss: 0.0064 - accuracy: 1.000 - ETA: 13s - loss: 0.0063 - accuracy: 1.000 - ETA: 13s - loss: 0.0066 - accuracy: 1.000 - ETA: 13s - loss: 0.0065 - accuracy: 1.000 - ETA: 13s - loss: 0.0062 - accuracy: 1.000 - ETA: 13s - loss: 0.0060 - accuracy: 1.000 - ETA: 12s - loss: 0.0061 - accuracy: 1.000 - ETA: 11s - loss: 0.0063 - accuracy: 1.000 - ETA: 11s - loss: 0.0062 - accuracy: 1.000 - ETA: 10s - loss: 0.0061 - accuracy: 1.000 - ETA: 9s - loss: 0.0061 - accuracy: 1.000 - ETA: 9s - loss: 0.0059 - accuracy: 1.00 - ETA: 8s - loss: 0.0059 - accuracy: 1.00 - ETA: 8s - loss: 0.0057 - accuracy: 1.00 - ETA: 7s - loss: 0.0057 - accuracy: 1.00 - ETA: 7s - loss: 0.0056 - accuracy: 1.00 - ETA: 6s - loss: 0.0058 - accuracy: 1.00 - ETA: 5s - loss: 0.0058 - accuracy: 1.00 - ETA: 5s - loss: 0.0057 - accuracy: 1.00 - ETA: 4s - loss: 0.0056 - accuracy: 1.00 - ETA: 3s - loss: 0.0055 - accuracy: 1.00 - ETA: 3s - loss: 0.0055 - accuracy: 1.00 - ETA: 2s - loss: 0.0054 - accuracy: 1.00 - ETA: 2s - loss: 0.0053 - accuracy: 1.00 - ETA: 1s - loss: 0.0053 - accuracy: 1.00 - ETA: 0s - loss: 0.0052 - accuracy: 1.00 - ETA: 0s - loss: 0.0052 - accuracy: 1.00 - 20s 19ms/step - loss: 0.0057 - accuracy: 1.0000\n",
      "Epoch 27/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0038 - accuracy: 1.000 - ETA: 17s - loss: 0.0047 - accuracy: 1.000 - ETA: 16s - loss: 0.0053 - accuracy: 1.000 - ETA: 16s - loss: 0.0052 - accuracy: 1.000 - ETA: 18s - loss: 0.0053 - accuracy: 1.000 - ETA: 17s - loss: 0.0054 - accuracy: 1.000 - ETA: 16s - loss: 0.0064 - accuracy: 1.000 - ETA: 15s - loss: 0.0065 - accuracy: 1.000 - ETA: 14s - loss: 0.0061 - accuracy: 1.000 - ETA: 13s - loss: 0.0075 - accuracy: 0.999 - ETA: 12s - loss: 0.0077 - accuracy: 0.999 - ETA: 11s - loss: 0.0075 - accuracy: 0.999 - ETA: 11s - loss: 0.0086 - accuracy: 0.999 - ETA: 11s - loss: 0.0089 - accuracy: 0.998 - ETA: 10s - loss: 0.0089 - accuracy: 0.998 - ETA: 9s - loss: 0.0092 - accuracy: 0.998 - ETA: 9s - loss: 0.0090 - accuracy: 0.99 - ETA: 8s - loss: 0.0109 - accuracy: 0.99 - ETA: 7s - loss: 0.0114 - accuracy: 0.99 - ETA: 7s - loss: 0.0114 - accuracy: 0.99 - ETA: 6s - loss: 0.0115 - accuracy: 0.99 - ETA: 5s - loss: 0.0115 - accuracy: 0.99 - ETA: 5s - loss: 0.0127 - accuracy: 0.99 - ETA: 4s - loss: 0.0136 - accuracy: 0.99 - ETA: 4s - loss: 0.0140 - accuracy: 0.99 - ETA: 3s - loss: 0.0151 - accuracy: 0.99 - ETA: 2s - loss: 0.0158 - accuracy: 0.99 - ETA: 2s - loss: 0.0160 - accuracy: 0.99 - ETA: 1s - loss: 0.0161 - accuracy: 0.99 - ETA: 0s - loss: 0.0166 - accuracy: 0.99 - ETA: 0s - loss: 0.0168 - accuracy: 0.99 - 20s 19ms/step - loss: 0.0174 - accuracy: 0.9949\n",
      "Epoch 28/50\n",
      "1008/1008 [==============================] - ETA: 30s - loss: 0.0197 - accuracy: 1.000 - ETA: 26s - loss: 0.0247 - accuracy: 0.995 - ETA: 22s - loss: 0.0239 - accuracy: 0.995 - ETA: 20s - loss: 0.0211 - accuracy: 0.996 - ETA: 19s - loss: 0.0283 - accuracy: 0.992 - ETA: 17s - loss: 0.0266 - accuracy: 0.994 - ETA: 16s - loss: 0.0351 - accuracy: 0.991 - ETA: 15s - loss: 0.0369 - accuracy: 0.990 - ETA: 14s - loss: 0.0389 - accuracy: 0.989 - ETA: 14s - loss: 0.0371 - accuracy: 0.989 - ETA: 14s - loss: 0.0394 - accuracy: 0.988 - ETA: 13s - loss: 0.0391 - accuracy: 0.988 - ETA: 12s - loss: 0.0392 - accuracy: 0.988 - ETA: 11s - loss: 0.0469 - accuracy: 0.985 - ETA: 10s - loss: 0.0447 - accuracy: 0.986 - ETA: 10s - loss: 0.0470 - accuracy: 0.984 - ETA: 9s - loss: 0.0495 - accuracy: 0.983 - ETA: 8s - loss: 0.0539 - accuracy: 0.98 - ETA: 7s - loss: 0.0538 - accuracy: 0.98 - ETA: 7s - loss: 0.0579 - accuracy: 0.97 - ETA: 6s - loss: 0.0581 - accuracy: 0.97 - ETA: 6s - loss: 0.0598 - accuracy: 0.97 - ETA: 5s - loss: 0.0603 - accuracy: 0.97 - ETA: 4s - loss: 0.0642 - accuracy: 0.97 - ETA: 4s - loss: 0.0648 - accuracy: 0.97 - ETA: 3s - loss: 0.0659 - accuracy: 0.97 - ETA: 2s - loss: 0.0670 - accuracy: 0.97 - ETA: 2s - loss: 0.0690 - accuracy: 0.97 - ETA: 1s - loss: 0.0712 - accuracy: 0.97 - ETA: 0s - loss: 0.0724 - accuracy: 0.97 - ETA: 0s - loss: 0.0727 - accuracy: 0.97 - 20s 20ms/step - loss: 0.0738 - accuracy: 0.9726\n",
      "Epoch 29/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.1098 - accuracy: 0.964 - ETA: 16s - loss: 0.0894 - accuracy: 0.957 - ETA: 16s - loss: 0.1051 - accuracy: 0.947 - ETA: 15s - loss: 0.1067 - accuracy: 0.946 - ETA: 15s - loss: 0.0988 - accuracy: 0.951 - ETA: 14s - loss: 0.0984 - accuracy: 0.953 - ETA: 13s - loss: 0.1010 - accuracy: 0.956 - ETA: 13s - loss: 0.1013 - accuracy: 0.957 - ETA: 14s - loss: 0.0956 - accuracy: 0.960 - ETA: 13s - loss: 0.0955 - accuracy: 0.959 - ETA: 12s - loss: 0.0973 - accuracy: 0.959 - ETA: 11s - loss: 0.0950 - accuracy: 0.960 - ETA: 11s - loss: 0.0941 - accuracy: 0.960 - ETA: 10s - loss: 0.0949 - accuracy: 0.960 - ETA: 9s - loss: 0.0982 - accuracy: 0.958 - ETA: 9s - loss: 0.0969 - accuracy: 0.95 - ETA: 8s - loss: 0.0961 - accuracy: 0.95 - ETA: 8s - loss: 0.0980 - accuracy: 0.95 - ETA: 7s - loss: 0.0973 - accuracy: 0.95 - ETA: 7s - loss: 0.0978 - accuracy: 0.95 - ETA: 6s - loss: 0.0974 - accuracy: 0.95 - ETA: 5s - loss: 0.0990 - accuracy: 0.95 - ETA: 5s - loss: 0.1026 - accuracy: 0.95 - ETA: 4s - loss: 0.1024 - accuracy: 0.95 - ETA: 3s - loss: 0.1009 - accuracy: 0.95 - ETA: 3s - loss: 0.1009 - accuracy: 0.95 - ETA: 2s - loss: 0.0987 - accuracy: 0.95 - ETA: 2s - loss: 0.1008 - accuracy: 0.95 - ETA: 1s - loss: 0.1000 - accuracy: 0.96 - ETA: 0s - loss: 0.1037 - accuracy: 0.95 - ETA: 0s - loss: 0.1038 - accuracy: 0.95 - 19s 19ms/step - loss: 0.1031 - accuracy: 0.9586\n",
      "Epoch 30/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0688 - accuracy: 0.977 - ETA: 16s - loss: 0.0730 - accuracy: 0.982 - ETA: 15s - loss: 0.0783 - accuracy: 0.976 - ETA: 15s - loss: 0.0741 - accuracy: 0.978 - ETA: 17s - loss: 0.0838 - accuracy: 0.976 - ETA: 16s - loss: 0.0810 - accuracy: 0.975 - ETA: 16s - loss: 0.0744 - accuracy: 0.977 - ETA: 15s - loss: 0.0779 - accuracy: 0.974 - ETA: 14s - loss: 0.0837 - accuracy: 0.971 - ETA: 13s - loss: 0.0786 - accuracy: 0.974 - ETA: 12s - loss: 0.0831 - accuracy: 0.971 - ETA: 11s - loss: 0.0803 - accuracy: 0.972 - ETA: 11s - loss: 0.0772 - accuracy: 0.974 - ETA: 11s - loss: 0.0782 - accuracy: 0.973 - ETA: 10s - loss: 0.0774 - accuracy: 0.973 - ETA: 9s - loss: 0.0761 - accuracy: 0.973 - ETA: 9s - loss: 0.0767 - accuracy: 0.97 - ETA: 8s - loss: 0.0756 - accuracy: 0.97 - ETA: 7s - loss: 0.0737 - accuracy: 0.97 - ETA: 7s - loss: 0.0729 - accuracy: 0.97 - ETA: 6s - loss: 0.0722 - accuracy: 0.97 - ETA: 5s - loss: 0.0722 - accuracy: 0.97 - ETA: 5s - loss: 0.0707 - accuracy: 0.97 - ETA: 4s - loss: 0.0688 - accuracy: 0.97 - ETA: 4s - loss: 0.0682 - accuracy: 0.97 - ETA: 3s - loss: 0.0682 - accuracy: 0.97 - ETA: 2s - loss: 0.0684 - accuracy: 0.97 - ETA: 2s - loss: 0.0692 - accuracy: 0.97 - ETA: 1s - loss: 0.0683 - accuracy: 0.97 - ETA: 0s - loss: 0.0674 - accuracy: 0.97 - ETA: 0s - loss: 0.0667 - accuracy: 0.97 - 20s 20ms/step - loss: 0.0668 - accuracy: 0.9782\n",
      "Epoch 31/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0404 - accuracy: 0.982 - ETA: 16s - loss: 0.0432 - accuracy: 0.982 - ETA: 16s - loss: 0.0400 - accuracy: 0.985 - ETA: 15s - loss: 0.0383 - accuracy: 0.986 - ETA: 15s - loss: 0.0356 - accuracy: 0.987 - ETA: 14s - loss: 0.0342 - accuracy: 0.988 - ETA: 14s - loss: 0.0333 - accuracy: 0.989 - ETA: 13s - loss: 0.0406 - accuracy: 0.987 - ETA: 13s - loss: 0.0399 - accuracy: 0.987 - ETA: 13s - loss: 0.0376 - accuracy: 0.988 - ETA: 13s - loss: 0.0358 - accuracy: 0.989 - ETA: 12s - loss: 0.0345 - accuracy: 0.990 - ETA: 11s - loss: 0.0349 - accuracy: 0.990 - ETA: 10s - loss: 0.0350 - accuracy: 0.990 - ETA: 10s - loss: 0.0360 - accuracy: 0.989 - ETA: 9s - loss: 0.0370 - accuracy: 0.988 - ETA: 8s - loss: 0.0366 - accuracy: 0.98 - ETA: 8s - loss: 0.0365 - accuracy: 0.98 - ETA: 8s - loss: 0.0365 - accuracy: 0.98 - ETA: 7s - loss: 0.0367 - accuracy: 0.98 - ETA: 6s - loss: 0.0369 - accuracy: 0.98 - ETA: 6s - loss: 0.0362 - accuracy: 0.98 - ETA: 5s - loss: 0.0358 - accuracy: 0.98 - ETA: 4s - loss: 0.0376 - accuracy: 0.98 - ETA: 4s - loss: 0.0373 - accuracy: 0.98 - ETA: 3s - loss: 0.0375 - accuracy: 0.98 - ETA: 2s - loss: 0.0371 - accuracy: 0.98 - ETA: 2s - loss: 0.0362 - accuracy: 0.98 - ETA: 1s - loss: 0.0361 - accuracy: 0.98 - ETA: 0s - loss: 0.0361 - accuracy: 0.98 - ETA: 0s - loss: 0.0355 - accuracy: 0.98 - 20s 20ms/step - loss: 0.0352 - accuracy: 0.9897\n",
      "Epoch 32/50\n",
      "1008/1008 [==============================] - ETA: 18s - loss: 0.0256 - accuracy: 0.991 - ETA: 18s - loss: 0.0280 - accuracy: 0.988 - ETA: 17s - loss: 0.0276 - accuracy: 0.992 - ETA: 17s - loss: 0.0269 - accuracy: 0.992 - ETA: 16s - loss: 0.0252 - accuracy: 0.993 - ETA: 17s - loss: 0.0260 - accuracy: 0.993 - ETA: 17s - loss: 0.0236 - accuracy: 0.994 - ETA: 16s - loss: 0.0221 - accuracy: 0.995 - ETA: 15s - loss: 0.0226 - accuracy: 0.994 - ETA: 14s - loss: 0.0245 - accuracy: 0.993 - ETA: 14s - loss: 0.0245 - accuracy: 0.993 - ETA: 13s - loss: 0.0238 - accuracy: 0.993 - ETA: 12s - loss: 0.0226 - accuracy: 0.994 - ETA: 11s - loss: 0.0227 - accuracy: 0.993 - ETA: 10s - loss: 0.0230 - accuracy: 0.993 - ETA: 10s - loss: 0.0225 - accuracy: 0.993 - ETA: 9s - loss: 0.0216 - accuracy: 0.994 - ETA: 9s - loss: 0.0211 - accuracy: 0.99 - ETA: 8s - loss: 0.0205 - accuracy: 0.99 - ETA: 7s - loss: 0.0208 - accuracy: 0.99 - ETA: 7s - loss: 0.0211 - accuracy: 0.99 - ETA: 6s - loss: 0.0214 - accuracy: 0.99 - ETA: 5s - loss: 0.0211 - accuracy: 0.99 - ETA: 5s - loss: 0.0205 - accuracy: 0.99 - ETA: 4s - loss: 0.0210 - accuracy: 0.99 - ETA: 3s - loss: 0.0206 - accuracy: 0.99 - ETA: 3s - loss: 0.0208 - accuracy: 0.99 - ETA: 2s - loss: 0.0208 - accuracy: 0.99 - ETA: 1s - loss: 0.0213 - accuracy: 0.99 - ETA: 1s - loss: 0.0215 - accuracy: 0.99 - ETA: 0s - loss: 0.0222 - accuracy: 0.99 - 21s 21ms/step - loss: 0.0221 - accuracy: 0.9949\n",
      "Epoch 33/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0086 - accuracy: 1.000 - ETA: 15s - loss: 0.0114 - accuracy: 1.000 - ETA: 15s - loss: 0.0118 - accuracy: 1.000 - ETA: 19s - loss: 0.0116 - accuracy: 1.000 - ETA: 18s - loss: 0.0116 - accuracy: 1.000 - ETA: 17s - loss: 0.0118 - accuracy: 1.000 - ETA: 16s - loss: 0.0118 - accuracy: 1.000 - ETA: 15s - loss: 0.0123 - accuracy: 1.000 - ETA: 14s - loss: 0.0125 - accuracy: 0.999 - ETA: 13s - loss: 0.0121 - accuracy: 0.999 - ETA: 12s - loss: 0.0123 - accuracy: 0.999 - ETA: 12s - loss: 0.0132 - accuracy: 0.998 - ETA: 12s - loss: 0.0142 - accuracy: 0.998 - ETA: 11s - loss: 0.0137 - accuracy: 0.998 - ETA: 10s - loss: 0.0133 - accuracy: 0.998 - ETA: 10s - loss: 0.0144 - accuracy: 0.998 - ETA: 9s - loss: 0.0145 - accuracy: 0.998 - ETA: 8s - loss: 0.0152 - accuracy: 0.99 - ETA: 7s - loss: 0.0149 - accuracy: 0.99 - ETA: 7s - loss: 0.0146 - accuracy: 0.99 - ETA: 6s - loss: 0.0143 - accuracy: 0.99 - ETA: 6s - loss: 0.0142 - accuracy: 0.99 - ETA: 5s - loss: 0.0140 - accuracy: 0.99 - ETA: 4s - loss: 0.0136 - accuracy: 0.99 - ETA: 4s - loss: 0.0135 - accuracy: 0.99 - ETA: 3s - loss: 0.0137 - accuracy: 0.99 - ETA: 2s - loss: 0.0142 - accuracy: 0.99 - ETA: 2s - loss: 0.0148 - accuracy: 0.99 - ETA: 1s - loss: 0.0149 - accuracy: 0.99 - ETA: 0s - loss: 0.0145 - accuracy: 0.99 - ETA: 0s - loss: 0.0146 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0147 - accuracy: 0.9972\n",
      "Epoch 34/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0089 - accuracy: 1.000 - ETA: 17s - loss: 0.0074 - accuracy: 1.000 - ETA: 16s - loss: 0.0088 - accuracy: 1.000 - ETA: 15s - loss: 0.0075 - accuracy: 1.000 - ETA: 15s - loss: 0.0095 - accuracy: 1.000 - ETA: 14s - loss: 0.0093 - accuracy: 1.000 - ETA: 14s - loss: 0.0094 - accuracy: 1.000 - ETA: 13s - loss: 0.0098 - accuracy: 1.000 - ETA: 14s - loss: 0.0093 - accuracy: 1.000 - ETA: 13s - loss: 0.0091 - accuracy: 1.000 - ETA: 13s - loss: 0.0091 - accuracy: 1.000 - ETA: 12s - loss: 0.0096 - accuracy: 1.000 - ETA: 11s - loss: 0.0095 - accuracy: 1.000 - ETA: 10s - loss: 0.0098 - accuracy: 0.999 - ETA: 10s - loss: 0.0098 - accuracy: 0.999 - ETA: 9s - loss: 0.0116 - accuracy: 0.998 - ETA: 8s - loss: 0.0115 - accuracy: 0.99 - ETA: 8s - loss: 0.0114 - accuracy: 0.99 - ETA: 7s - loss: 0.0114 - accuracy: 0.99 - ETA: 7s - loss: 0.0110 - accuracy: 0.99 - ETA: 6s - loss: 0.0107 - accuracy: 0.99 - ETA: 5s - loss: 0.0107 - accuracy: 0.99 - ETA: 5s - loss: 0.0104 - accuracy: 0.99 - ETA: 4s - loss: 0.0102 - accuracy: 0.99 - ETA: 4s - loss: 0.0100 - accuracy: 0.99 - ETA: 3s - loss: 0.0099 - accuracy: 0.99 - ETA: 2s - loss: 0.0098 - accuracy: 0.99 - ETA: 2s - loss: 0.0097 - accuracy: 0.99 - ETA: 1s - loss: 0.0098 - accuracy: 0.99 - ETA: 0s - loss: 0.0103 - accuracy: 0.99 - ETA: 0s - loss: 0.0108 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0107 - accuracy: 0.9987\n",
      "Epoch 35/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0055 - accuracy: 1.000 - ETA: 16s - loss: 0.0047 - accuracy: 1.000 - ETA: 16s - loss: 0.0057 - accuracy: 1.000 - ETA: 15s - loss: 0.0047 - accuracy: 1.000 - ETA: 15s - loss: 0.0045 - accuracy: 1.000 - ETA: 14s - loss: 0.0066 - accuracy: 0.998 - ETA: 13s - loss: 0.0066 - accuracy: 0.998 - ETA: 13s - loss: 0.0062 - accuracy: 0.998 - ETA: 13s - loss: 0.0065 - accuracy: 0.999 - ETA: 13s - loss: 0.0063 - accuracy: 0.999 - ETA: 12s - loss: 0.0061 - accuracy: 0.999 - ETA: 12s - loss: 0.0059 - accuracy: 0.999 - ETA: 11s - loss: 0.0061 - accuracy: 0.999 - ETA: 10s - loss: 0.0061 - accuracy: 0.999 - ETA: 10s - loss: 0.0060 - accuracy: 0.999 - ETA: 9s - loss: 0.0059 - accuracy: 0.999 - ETA: 8s - loss: 0.0061 - accuracy: 0.99 - ETA: 8s - loss: 0.0059 - accuracy: 0.99 - ETA: 7s - loss: 0.0059 - accuracy: 0.99 - ETA: 7s - loss: 0.0059 - accuracy: 0.99 - ETA: 6s - loss: 0.0057 - accuracy: 0.99 - ETA: 5s - loss: 0.0059 - accuracy: 0.99 - ETA: 5s - loss: 0.0061 - accuracy: 0.99 - ETA: 4s - loss: 0.0060 - accuracy: 0.99 - ETA: 4s - loss: 0.0061 - accuracy: 0.99 - ETA: 3s - loss: 0.0061 - accuracy: 0.99 - ETA: 2s - loss: 0.0062 - accuracy: 0.99 - ETA: 2s - loss: 0.0063 - accuracy: 0.99 - ETA: 1s - loss: 0.0062 - accuracy: 0.99 - ETA: 0s - loss: 0.0061 - accuracy: 0.99 - ETA: 0s - loss: 0.0062 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0062 - accuracy: 0.9996\n",
      "Epoch 36/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0042 - accuracy: 1.000 - ETA: 17s - loss: 0.0038 - accuracy: 1.000 - ETA: 16s - loss: 0.0047 - accuracy: 1.000 - ETA: 16s - loss: 0.0081 - accuracy: 0.997 - ETA: 15s - loss: 0.0069 - accuracy: 0.998 - ETA: 14s - loss: 0.0063 - accuracy: 0.998 - ETA: 13s - loss: 0.0065 - accuracy: 0.998 - ETA: 13s - loss: 0.0063 - accuracy: 0.998 - ETA: 12s - loss: 0.0058 - accuracy: 0.999 - ETA: 13s - loss: 0.0057 - accuracy: 0.999 - ETA: 12s - loss: 0.0059 - accuracy: 0.999 - ETA: 12s - loss: 0.0057 - accuracy: 0.999 - ETA: 11s - loss: 0.0056 - accuracy: 0.999 - ETA: 10s - loss: 0.0057 - accuracy: 0.999 - ETA: 10s - loss: 0.0057 - accuracy: 0.999 - ETA: 9s - loss: 0.0055 - accuracy: 0.999 - ETA: 8s - loss: 0.0058 - accuracy: 0.99 - ETA: 8s - loss: 0.0057 - accuracy: 0.99 - ETA: 7s - loss: 0.0056 - accuracy: 0.99 - ETA: 6s - loss: 0.0054 - accuracy: 0.99 - ETA: 6s - loss: 0.0053 - accuracy: 0.99 - ETA: 5s - loss: 0.0051 - accuracy: 0.99 - ETA: 5s - loss: 0.0053 - accuracy: 0.99 - ETA: 4s - loss: 0.0052 - accuracy: 0.99 - ETA: 4s - loss: 0.0050 - accuracy: 0.99 - ETA: 3s - loss: 0.0050 - accuracy: 0.99 - ETA: 2s - loss: 0.0050 - accuracy: 0.99 - ETA: 2s - loss: 0.0049 - accuracy: 0.99 - ETA: 1s - loss: 0.0048 - accuracy: 0.99 - ETA: 0s - loss: 0.0047 - accuracy: 0.99 - ETA: 0s - loss: 0.0047 - accuracy: 0.99 - 20s 20ms/step - loss: 0.0049 - accuracy: 0.9996\n",
      "Epoch 37/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0042 - accuracy: 1.000 - ETA: 17s - loss: 0.0045 - accuracy: 1.000 - ETA: 17s - loss: 0.0049 - accuracy: 1.000 - ETA: 16s - loss: 0.0041 - accuracy: 1.000 - ETA: 15s - loss: 0.0041 - accuracy: 1.000 - ETA: 14s - loss: 0.0038 - accuracy: 1.000 - ETA: 15s - loss: 0.0041 - accuracy: 1.000 - ETA: 15s - loss: 0.0040 - accuracy: 1.000 - ETA: 14s - loss: 0.0047 - accuracy: 1.000 - ETA: 13s - loss: 0.0045 - accuracy: 1.000 - ETA: 13s - loss: 0.0044 - accuracy: 1.000 - ETA: 12s - loss: 0.0047 - accuracy: 1.000 - ETA: 11s - loss: 0.0045 - accuracy: 1.000 - ETA: 10s - loss: 0.0045 - accuracy: 1.000 - ETA: 10s - loss: 0.0045 - accuracy: 1.000 - ETA: 9s - loss: 0.0049 - accuracy: 1.000 - ETA: 9s - loss: 0.0047 - accuracy: 1.00 - ETA: 8s - loss: 0.0046 - accuracy: 1.00 - ETA: 8s - loss: 0.0045 - accuracy: 1.00 - ETA: 7s - loss: 0.0045 - accuracy: 1.00 - ETA: 6s - loss: 0.0044 - accuracy: 1.00 - ETA: 5s - loss: 0.0045 - accuracy: 1.00 - ETA: 5s - loss: 0.0044 - accuracy: 1.00 - ETA: 4s - loss: 0.0044 - accuracy: 1.00 - ETA: 4s - loss: 0.0043 - accuracy: 1.00 - ETA: 3s - loss: 0.0042 - accuracy: 1.00 - ETA: 2s - loss: 0.0043 - accuracy: 1.00 - ETA: 2s - loss: 0.0042 - accuracy: 1.00 - ETA: 1s - loss: 0.0042 - accuracy: 1.00 - ETA: 0s - loss: 0.0041 - accuracy: 1.00 - ETA: 0s - loss: 0.0041 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0041 - accuracy: 1.0000\n",
      "Epoch 38/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0016 - accuracy: 1.000 - ETA: 17s - loss: 0.0016 - accuracy: 1.000 - ETA: 16s - loss: 0.0024 - accuracy: 1.000 - ETA: 15s - loss: 0.0029 - accuracy: 1.000 - ETA: 15s - loss: 0.0033 - accuracy: 1.000 - ETA: 15s - loss: 0.0039 - accuracy: 1.000 - ETA: 16s - loss: 0.0036 - accuracy: 1.000 - ETA: 15s - loss: 0.0035 - accuracy: 1.000 - ETA: 14s - loss: 0.0032 - accuracy: 1.000 - ETA: 13s - loss: 0.0031 - accuracy: 1.000 - ETA: 12s - loss: 0.0030 - accuracy: 1.000 - ETA: 12s - loss: 0.0032 - accuracy: 1.000 - ETA: 11s - loss: 0.0034 - accuracy: 1.000 - ETA: 10s - loss: 0.0034 - accuracy: 1.000 - ETA: 10s - loss: 0.0033 - accuracy: 1.000 - ETA: 9s - loss: 0.0034 - accuracy: 1.000 - ETA: 9s - loss: 0.0034 - accuracy: 1.00 - ETA: 8s - loss: 0.0035 - accuracy: 1.00 - ETA: 8s - loss: 0.0036 - accuracy: 1.00 - ETA: 7s - loss: 0.0037 - accuracy: 1.00 - ETA: 6s - loss: 0.0037 - accuracy: 1.00 - ETA: 6s - loss: 0.0037 - accuracy: 1.00 - ETA: 5s - loss: 0.0036 - accuracy: 1.00 - ETA: 4s - loss: 0.0037 - accuracy: 1.00 - ETA: 4s - loss: 0.0037 - accuracy: 1.00 - ETA: 3s - loss: 0.0037 - accuracy: 1.00 - ETA: 2s - loss: 0.0037 - accuracy: 1.00 - ETA: 2s - loss: 0.0037 - accuracy: 1.00 - ETA: 1s - loss: 0.0036 - accuracy: 1.00 - ETA: 0s - loss: 0.0036 - accuracy: 1.00 - ETA: 0s - loss: 0.0035 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0036 - accuracy: 1.0000\n",
      "Epoch 39/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0064 - accuracy: 1.000 - ETA: 16s - loss: 0.0050 - accuracy: 1.000 - ETA: 15s - loss: 0.0048 - accuracy: 1.000 - ETA: 15s - loss: 0.0040 - accuracy: 1.000 - ETA: 15s - loss: 0.0046 - accuracy: 1.000 - ETA: 16s - loss: 0.0054 - accuracy: 1.000 - ETA: 16s - loss: 0.0047 - accuracy: 1.000 - ETA: 15s - loss: 0.0046 - accuracy: 1.000 - ETA: 14s - loss: 0.0043 - accuracy: 1.000 - ETA: 13s - loss: 0.0042 - accuracy: 1.000 - ETA: 12s - loss: 0.0039 - accuracy: 1.000 - ETA: 12s - loss: 0.0036 - accuracy: 1.000 - ETA: 11s - loss: 0.0035 - accuracy: 1.000 - ETA: 10s - loss: 0.0035 - accuracy: 1.000 - ETA: 10s - loss: 0.0035 - accuracy: 1.000 - ETA: 10s - loss: 0.0034 - accuracy: 1.000 - ETA: 9s - loss: 0.0033 - accuracy: 1.000 - ETA: 8s - loss: 0.0032 - accuracy: 1.00 - ETA: 8s - loss: 0.0031 - accuracy: 1.00 - ETA: 7s - loss: 0.0031 - accuracy: 1.00 - ETA: 6s - loss: 0.0030 - accuracy: 1.00 - ETA: 6s - loss: 0.0032 - accuracy: 1.00 - ETA: 5s - loss: 0.0031 - accuracy: 1.00 - ETA: 4s - loss: 0.0031 - accuracy: 1.00 - ETA: 4s - loss: 0.0031 - accuracy: 1.00 - ETA: 3s - loss: 0.0031 - accuracy: 1.00 - ETA: 2s - loss: 0.0032 - accuracy: 1.00 - ETA: 2s - loss: 0.0033 - accuracy: 1.00 - ETA: 1s - loss: 0.0033 - accuracy: 1.00 - ETA: 0s - loss: 0.0034 - accuracy: 1.00 - ETA: 0s - loss: 0.0035 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0035 - accuracy: 1.0000\n",
      "Epoch 40/50\n",
      "1008/1008 [==============================] - ETA: 19s - loss: 0.0016 - accuracy: 1.000 - ETA: 19s - loss: 0.0015 - accuracy: 1.000 - ETA: 22s - loss: 0.0016 - accuracy: 1.000 - ETA: 20s - loss: 0.0017 - accuracy: 1.000 - ETA: 19s - loss: 0.0016 - accuracy: 1.000 - ETA: 17s - loss: 0.0017 - accuracy: 1.000 - ETA: 16s - loss: 0.0018 - accuracy: 1.000 - ETA: 16s - loss: 0.0018 - accuracy: 1.000 - ETA: 15s - loss: 0.0019 - accuracy: 1.000 - ETA: 14s - loss: 0.0018 - accuracy: 1.000 - ETA: 14s - loss: 0.0018 - accuracy: 1.000 - ETA: 13s - loss: 0.0020 - accuracy: 1.000 - ETA: 12s - loss: 0.0020 - accuracy: 1.000 - ETA: 12s - loss: 0.0020 - accuracy: 1.000 - ETA: 11s - loss: 0.0020 - accuracy: 1.000 - ETA: 10s - loss: 0.0020 - accuracy: 1.000 - ETA: 9s - loss: 0.0020 - accuracy: 1.000 - ETA: 9s - loss: 0.0019 - accuracy: 1.00 - ETA: 8s - loss: 0.0019 - accuracy: 1.00 - ETA: 8s - loss: 0.0019 - accuracy: 1.00 - ETA: 7s - loss: 0.0019 - accuracy: 1.00 - ETA: 6s - loss: 0.0021 - accuracy: 1.00 - ETA: 5s - loss: 0.0022 - accuracy: 1.00 - ETA: 5s - loss: 0.0022 - accuracy: 1.00 - ETA: 4s - loss: 0.0022 - accuracy: 1.00 - ETA: 3s - loss: 0.0022 - accuracy: 1.00 - ETA: 3s - loss: 0.0021 - accuracy: 1.00 - ETA: 2s - loss: 0.0021 - accuracy: 1.00 - ETA: 1s - loss: 0.0021 - accuracy: 1.00 - ETA: 1s - loss: 0.0021 - accuracy: 1.00 - ETA: 0s - loss: 0.0021 - accuracy: 1.00 - 22s 22ms/step - loss: 0.0021 - accuracy: 1.0000\n",
      "Epoch 41/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0025 - accuracy: 1.000 - ETA: 19s - loss: 0.0095 - accuracy: 0.997 - ETA: 18s - loss: 0.0074 - accuracy: 0.998 - ETA: 17s - loss: 0.0059 - accuracy: 0.998 - ETA: 18s - loss: 0.0048 - accuracy: 0.999 - ETA: 18s - loss: 0.0045 - accuracy: 0.999 - ETA: 17s - loss: 0.0043 - accuracy: 0.999 - ETA: 16s - loss: 0.0039 - accuracy: 0.999 - ETA: 15s - loss: 0.0038 - accuracy: 0.999 - ETA: 14s - loss: 0.0036 - accuracy: 0.999 - ETA: 13s - loss: 0.0033 - accuracy: 0.999 - ETA: 12s - loss: 0.0033 - accuracy: 0.999 - ETA: 12s - loss: 0.0031 - accuracy: 0.999 - ETA: 11s - loss: 0.0030 - accuracy: 0.999 - ETA: 11s - loss: 0.0029 - accuracy: 0.999 - ETA: 10s - loss: 0.0028 - accuracy: 0.999 - ETA: 9s - loss: 0.0027 - accuracy: 0.999 - ETA: 9s - loss: 0.0028 - accuracy: 0.99 - ETA: 8s - loss: 0.0027 - accuracy: 0.99 - ETA: 7s - loss: 0.0027 - accuracy: 0.99 - ETA: 6s - loss: 0.0026 - accuracy: 0.99 - ETA: 6s - loss: 0.0026 - accuracy: 0.99 - ETA: 5s - loss: 0.0032 - accuracy: 0.99 - ETA: 4s - loss: 0.0032 - accuracy: 0.99 - ETA: 4s - loss: 0.0032 - accuracy: 0.99 - ETA: 3s - loss: 0.0032 - accuracy: 0.99 - ETA: 3s - loss: 0.0033 - accuracy: 0.99 - ETA: 2s - loss: 0.0032 - accuracy: 0.99 - ETA: 1s - loss: 0.0034 - accuracy: 0.99 - ETA: 0s - loss: 0.0033 - accuracy: 0.99 - ETA: 0s - loss: 0.0032 - accuracy: 0.99 - 21s 21ms/step - loss: 0.0032 - accuracy: 0.9996\n",
      "Epoch 42/50\n",
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0043 - accuracy: 1.000 - ETA: 22s - loss: 0.0036 - accuracy: 1.000 - ETA: 22s - loss: 0.0034 - accuracy: 1.000 - ETA: 20s - loss: 0.0030 - accuracy: 1.000 - ETA: 19s - loss: 0.0029 - accuracy: 1.000 - ETA: 17s - loss: 0.0031 - accuracy: 1.000 - ETA: 16s - loss: 0.0030 - accuracy: 1.000 - ETA: 15s - loss: 0.0027 - accuracy: 1.000 - ETA: 14s - loss: 0.0025 - accuracy: 1.000 - ETA: 14s - loss: 0.0026 - accuracy: 1.000 - ETA: 14s - loss: 0.0025 - accuracy: 1.000 - ETA: 13s - loss: 0.0024 - accuracy: 1.000 - ETA: 12s - loss: 0.0023 - accuracy: 1.000 - ETA: 12s - loss: 0.0026 - accuracy: 1.000 - ETA: 11s - loss: 0.0026 - accuracy: 1.000 - ETA: 10s - loss: 0.0026 - accuracy: 1.000 - ETA: 9s - loss: 0.0026 - accuracy: 1.000 - ETA: 8s - loss: 0.0027 - accuracy: 1.00 - ETA: 8s - loss: 0.0027 - accuracy: 1.00 - ETA: 7s - loss: 0.0028 - accuracy: 1.00 - ETA: 7s - loss: 0.0027 - accuracy: 1.00 - ETA: 6s - loss: 0.0027 - accuracy: 1.00 - ETA: 5s - loss: 0.0027 - accuracy: 1.00 - ETA: 5s - loss: 0.0026 - accuracy: 1.00 - ETA: 4s - loss: 0.0026 - accuracy: 1.00 - ETA: 3s - loss: 0.0026 - accuracy: 1.00 - ETA: 3s - loss: 0.0026 - accuracy: 1.00 - ETA: 2s - loss: 0.0027 - accuracy: 1.00 - ETA: 1s - loss: 0.0027 - accuracy: 1.00 - ETA: 1s - loss: 0.0027 - accuracy: 1.00 - ETA: 0s - loss: 0.0027 - accuracy: 1.00 - 21s 21ms/step - loss: 0.0026 - accuracy: 1.0000\n",
      "Epoch 43/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 0.0036 - accuracy: 1.000 - ETA: 17s - loss: 0.0024 - accuracy: 1.000 - ETA: 16s - loss: 0.0022 - accuracy: 1.000 - ETA: 15s - loss: 0.0019 - accuracy: 1.000 - ETA: 16s - loss: 0.0019 - accuracy: 1.000 - ETA: 17s - loss: 0.0019 - accuracy: 1.000 - ETA: 16s - loss: 0.0018 - accuracy: 1.000 - ETA: 15s - loss: 0.0020 - accuracy: 1.000 - ETA: 14s - loss: 0.0020 - accuracy: 1.000 - ETA: 13s - loss: 0.0019 - accuracy: 1.000 - ETA: 13s - loss: 0.0019 - accuracy: 1.000 - ETA: 12s - loss: 0.0019 - accuracy: 1.000 - ETA: 11s - loss: 0.0018 - accuracy: 1.000 - ETA: 11s - loss: 0.0019 - accuracy: 1.000 - ETA: 11s - loss: 0.0019 - accuracy: 1.000 - ETA: 10s - loss: 0.0019 - accuracy: 1.000 - ETA: 9s - loss: 0.0020 - accuracy: 1.000 - ETA: 8s - loss: 0.0020 - accuracy: 1.00 - ETA: 8s - loss: 0.0020 - accuracy: 1.00 - ETA: 7s - loss: 0.0021 - accuracy: 1.00 - ETA: 6s - loss: 0.0021 - accuracy: 1.00 - ETA: 6s - loss: 0.0020 - accuracy: 1.00 - ETA: 5s - loss: 0.0020 - accuracy: 1.00 - ETA: 5s - loss: 0.0020 - accuracy: 1.00 - ETA: 4s - loss: 0.0020 - accuracy: 1.00 - ETA: 3s - loss: 0.0019 - accuracy: 1.00 - ETA: 2s - loss: 0.0020 - accuracy: 1.00 - ETA: 2s - loss: 0.0020 - accuracy: 1.00 - ETA: 1s - loss: 0.0020 - accuracy: 1.00 - ETA: 0s - loss: 0.0020 - accuracy: 1.00 - ETA: 0s - loss: 0.0020 - accuracy: 1.00 - 21s 20ms/step - loss: 0.0020 - accuracy: 1.0000\n",
      "Epoch 44/50\n",
      "1008/1008 [==============================] - ETA: 35s - loss: 8.4227e-04 - accuracy: 1.000 - ETA: 26s - loss: 9.8684e-04 - accuracy: 1.000 - ETA: 23s - loss: 0.0013 - accuracy: 1.0000    - ETA: 20s - loss: 0.0018 - accuracy: 1.000 - ETA: 19s - loss: 0.0016 - accuracy: 1.000 - ETA: 17s - loss: 0.0022 - accuracy: 1.000 - ETA: 16s - loss: 0.0020 - accuracy: 1.000 - ETA: 15s - loss: 0.0019 - accuracy: 1.000 - ETA: 15s - loss: 0.0021 - accuracy: 1.000 - ETA: 15s - loss: 0.0021 - accuracy: 1.000 - ETA: 14s - loss: 0.0020 - accuracy: 1.000 - ETA: 13s - loss: 0.0020 - accuracy: 1.000 - ETA: 12s - loss: 0.0021 - accuracy: 1.000 - ETA: 12s - loss: 0.0020 - accuracy: 1.000 - ETA: 11s - loss: 0.0021 - accuracy: 1.000 - ETA: 10s - loss: 0.0020 - accuracy: 1.000 - ETA: 9s - loss: 0.0019 - accuracy: 1.000 - ETA: 9s - loss: 0.0019 - accuracy: 1.00 - ETA: 8s - loss: 0.0019 - accuracy: 1.00 - ETA: 7s - loss: 0.0020 - accuracy: 1.00 - ETA: 7s - loss: 0.0019 - accuracy: 1.00 - ETA: 6s - loss: 0.0019 - accuracy: 1.00 - ETA: 5s - loss: 0.0018 - accuracy: 1.00 - ETA: 5s - loss: 0.0018 - accuracy: 1.00 - ETA: 4s - loss: 0.0017 - accuracy: 1.00 - ETA: 3s - loss: 0.0018 - accuracy: 1.00 - ETA: 3s - loss: 0.0017 - accuracy: 1.00 - ETA: 2s - loss: 0.0018 - accuracy: 1.00 - ETA: 1s - loss: 0.0017 - accuracy: 1.00 - ETA: 1s - loss: 0.0017 - accuracy: 1.00 - ETA: 0s - loss: 0.0017 - accuracy: 1.00 - 21s 21ms/step - loss: 0.0018 - accuracy: 1.0000\n",
      "Epoch 45/50\n",
      "1008/1008 [==============================] - ETA: 18s - loss: 5.9007e-04 - accuracy: 1.000 - ETA: 25s - loss: 0.0022 - accuracy: 1.0000    - ETA: 26s - loss: 0.0017 - accuracy: 1.000 - ETA: 23s - loss: 0.0014 - accuracy: 1.000 - ETA: 21s - loss: 0.0013 - accuracy: 1.000 - ETA: 19s - loss: 0.0013 - accuracy: 1.000 - ETA: 18s - loss: 0.0013 - accuracy: 1.000 - ETA: 17s - loss: 0.0013 - accuracy: 1.000 - ETA: 16s - loss: 0.0013 - accuracy: 1.000 - ETA: 15s - loss: 0.0014 - accuracy: 1.000 - ETA: 15s - loss: 0.0016 - accuracy: 1.000 - ETA: 14s - loss: 0.0016 - accuracy: 1.000 - ETA: 13s - loss: 0.0017 - accuracy: 1.000 - ETA: 12s - loss: 0.0019 - accuracy: 1.000 - ETA: 11s - loss: 0.0019 - accuracy: 1.000 - ETA: 10s - loss: 0.0018 - accuracy: 1.000 - ETA: 10s - loss: 0.0018 - accuracy: 1.000 - ETA: 9s - loss: 0.0017 - accuracy: 1.000 - ETA: 8s - loss: 0.0017 - accuracy: 1.00 - ETA: 8s - loss: 0.0017 - accuracy: 1.00 - ETA: 7s - loss: 0.0016 - accuracy: 1.00 - ETA: 6s - loss: 0.0017 - accuracy: 1.00 - ETA: 5s - loss: 0.0018 - accuracy: 1.00 - ETA: 5s - loss: 0.0018 - accuracy: 1.00 - ETA: 4s - loss: 0.0017 - accuracy: 1.00 - ETA: 3s - loss: 0.0019 - accuracy: 1.00 - ETA: 3s - loss: 0.0018 - accuracy: 1.00 - ETA: 2s - loss: 0.0018 - accuracy: 1.00 - ETA: 1s - loss: 0.0018 - accuracy: 1.00 - ETA: 1s - loss: 0.0018 - accuracy: 1.00 - ETA: 0s - loss: 0.0018 - accuracy: 1.00 - 22s 22ms/step - loss: 0.0018 - accuracy: 1.0000\n",
      "Epoch 46/50\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1008/1008 [==============================] - ETA: 17s - loss: 0.0024 - accuracy: 1.000 - ETA: 17s - loss: 0.0032 - accuracy: 1.000 - ETA: 16s - loss: 0.0026 - accuracy: 1.000 - ETA: 15s - loss: 0.0023 - accuracy: 1.000 - ETA: 15s - loss: 0.0021 - accuracy: 1.000 - ETA: 14s - loss: 0.0019 - accuracy: 1.000 - ETA: 15s - loss: 0.0018 - accuracy: 1.000 - ETA: 15s - loss: 0.0017 - accuracy: 1.000 - ETA: 14s - loss: 0.0017 - accuracy: 1.000 - ETA: 13s - loss: 0.0017 - accuracy: 1.000 - ETA: 13s - loss: 0.0017 - accuracy: 1.000 - ETA: 12s - loss: 0.0016 - accuracy: 1.000 - ETA: 11s - loss: 0.0017 - accuracy: 1.000 - ETA: 11s - loss: 0.0018 - accuracy: 1.000 - ETA: 10s - loss: 0.0019 - accuracy: 1.000 - ETA: 10s - loss: 0.0019 - accuracy: 1.000 - ETA: 9s - loss: 0.0018 - accuracy: 1.000 - ETA: 9s - loss: 0.0018 - accuracy: 1.00 - ETA: 8s - loss: 0.0017 - accuracy: 1.00 - ETA: 7s - loss: 0.0017 - accuracy: 1.00 - ETA: 7s - loss: 0.0016 - accuracy: 1.00 - ETA: 6s - loss: 0.0016 - accuracy: 1.00 - ETA: 5s - loss: 0.0016 - accuracy: 1.00 - ETA: 5s - loss: 0.0015 - accuracy: 1.00 - ETA: 4s - loss: 0.0015 - accuracy: 1.00 - ETA: 3s - loss: 0.0015 - accuracy: 1.00 - ETA: 3s - loss: 0.0014 - accuracy: 1.00 - ETA: 2s - loss: 0.0014 - accuracy: 1.00 - ETA: 1s - loss: 0.0015 - accuracy: 1.00 - ETA: 1s - loss: 0.0015 - accuracy: 1.00 - ETA: 0s - loss: 0.0015 - accuracy: 1.00 - 23s 23ms/step - loss: 0.0015 - accuracy: 1.0000\n",
      "Epoch 47/50\n",
      "1008/1008 [==============================] - ETA: 21s - loss: 0.0021 - accuracy: 1.000 - ETA: 21s - loss: 0.0020 - accuracy: 1.000 - ETA: 21s - loss: 0.0019 - accuracy: 1.000 - ETA: 20s - loss: 0.0031 - accuracy: 1.000 - ETA: 19s - loss: 0.0026 - accuracy: 1.000 - ETA: 18s - loss: 0.0024 - accuracy: 1.000 - ETA: 19s - loss: 0.0023 - accuracy: 1.000 - ETA: 19s - loss: 0.0022 - accuracy: 1.000 - ETA: 18s - loss: 0.0020 - accuracy: 1.000 - ETA: 16s - loss: 0.0019 - accuracy: 1.000 - ETA: 15s - loss: 0.0018 - accuracy: 1.000 - ETA: 14s - loss: 0.0017 - accuracy: 1.000 - ETA: 13s - loss: 0.0016 - accuracy: 1.000 - ETA: 13s - loss: 0.0016 - accuracy: 1.000 - ETA: 13s - loss: 0.0016 - accuracy: 1.000 - ETA: 12s - loss: 0.0015 - accuracy: 1.000 - ETA: 11s - loss: 0.0015 - accuracy: 1.000 - ETA: 10s - loss: 0.0015 - accuracy: 1.000 - ETA: 9s - loss: 0.0015 - accuracy: 1.000 - ETA: 8s - loss: 0.0017 - accuracy: 1.00 - ETA: 8s - loss: 0.0016 - accuracy: 1.00 - ETA: 7s - loss: 0.0016 - accuracy: 1.00 - ETA: 6s - loss: 0.0016 - accuracy: 1.00 - ETA: 5s - loss: 0.0016 - accuracy: 1.00 - ETA: 5s - loss: 0.0016 - accuracy: 1.00 - ETA: 4s - loss: 0.0015 - accuracy: 1.00 - ETA: 3s - loss: 0.0015 - accuracy: 1.00 - ETA: 2s - loss: 0.0015 - accuracy: 1.00 - ETA: 1s - loss: 0.0014 - accuracy: 1.00 - ETA: 1s - loss: 0.0014 - accuracy: 1.00 - ETA: 0s - loss: 0.0014 - accuracy: 1.00 - 25s 25ms/step - loss: 0.0014 - accuracy: 1.0000\n",
      "Epoch 48/50\n",
      "1008/1008 [==============================] - ETA: 21s - loss: 0.0013 - accuracy: 1.000 - ETA: 19s - loss: 9.2848e-04 - accuracy: 1.000 - ETA: 18s - loss: 0.0027 - accuracy: 1.0000    - ETA: 18s - loss: 0.0023 - accuracy: 1.000 - ETA: 17s - loss: 0.0019 - accuracy: 1.000 - ETA: 16s - loss: 0.0017 - accuracy: 1.000 - ETA: 19s - loss: 0.0019 - accuracy: 1.000 - ETA: 18s - loss: 0.0017 - accuracy: 1.000 - ETA: 17s - loss: 0.0016 - accuracy: 1.000 - ETA: 16s - loss: 0.0015 - accuracy: 1.000 - ETA: 15s - loss: 0.0015 - accuracy: 1.000 - ETA: 14s - loss: 0.0015 - accuracy: 1.000 - ETA: 14s - loss: 0.0016 - accuracy: 1.000 - ETA: 13s - loss: 0.0016 - accuracy: 1.000 - ETA: 12s - loss: 0.0018 - accuracy: 1.000 - ETA: 12s - loss: 0.0017 - accuracy: 1.000 - ETA: 11s - loss: 0.0018 - accuracy: 1.000 - ETA: 10s - loss: 0.0018 - accuracy: 1.000 - ETA: 9s - loss: 0.0017 - accuracy: 1.000 - ETA: 8s - loss: 0.0017 - accuracy: 1.00 - ETA: 7s - loss: 0.0019 - accuracy: 1.00 - ETA: 7s - loss: 0.0018 - accuracy: 1.00 - ETA: 6s - loss: 0.0018 - accuracy: 1.00 - ETA: 5s - loss: 0.0017 - accuracy: 1.00 - ETA: 4s - loss: 0.0017 - accuracy: 1.00 - ETA: 4s - loss: 0.0017 - accuracy: 1.00 - ETA: 3s - loss: 0.0017 - accuracy: 1.00 - ETA: 2s - loss: 0.0017 - accuracy: 1.00 - ETA: 1s - loss: 0.0016 - accuracy: 1.00 - ETA: 1s - loss: 0.0016 - accuracy: 1.00 - ETA: 0s - loss: 0.0016 - accuracy: 1.00 - 23s 23ms/step - loss: 0.0016 - accuracy: 1.0000\n",
      "Epoch 49/50\n",
      "1008/1008 [==============================] - ETA: 19s - loss: 0.0044 - accuracy: 1.000 - ETA: 18s - loss: 0.0027 - accuracy: 1.000 - ETA: 17s - loss: 0.0022 - accuracy: 1.000 - ETA: 16s - loss: 0.0024 - accuracy: 1.000 - ETA: 15s - loss: 0.0021 - accuracy: 1.000 - ETA: 14s - loss: 0.0022 - accuracy: 1.000 - ETA: 15s - loss: 0.0019 - accuracy: 1.000 - ETA: 15s - loss: 0.0020 - accuracy: 1.000 - ETA: 14s - loss: 0.0019 - accuracy: 1.000 - ETA: 14s - loss: 0.0018 - accuracy: 1.000 - ETA: 13s - loss: 0.0017 - accuracy: 1.000 - ETA: 12s - loss: 0.0017 - accuracy: 1.000 - ETA: 11s - loss: 0.0017 - accuracy: 1.000 - ETA: 10s - loss: 0.0016 - accuracy: 1.000 - ETA: 10s - loss: 0.0017 - accuracy: 1.000 - ETA: 9s - loss: 0.0017 - accuracy: 1.000 - ETA: 9s - loss: 0.0017 - accuracy: 1.00 - ETA: 8s - loss: 0.0016 - accuracy: 1.00 - ETA: 7s - loss: 0.0016 - accuracy: 1.00 - ETA: 7s - loss: 0.0016 - accuracy: 1.00 - ETA: 6s - loss: 0.0015 - accuracy: 1.00 - ETA: 5s - loss: 0.0015 - accuracy: 1.00 - ETA: 5s - loss: 0.0015 - accuracy: 1.00 - ETA: 4s - loss: 0.0015 - accuracy: 1.00 - ETA: 4s - loss: 0.0015 - accuracy: 1.00 - ETA: 3s - loss: 0.0015 - accuracy: 1.00 - ETA: 2s - loss: 0.0015 - accuracy: 1.00 - ETA: 2s - loss: 0.0014 - accuracy: 1.00 - ETA: 1s - loss: 0.0014 - accuracy: 1.00 - ETA: 0s - loss: 0.0014 - accuracy: 1.00 - ETA: 0s - loss: 0.0014 - accuracy: 1.00 - 20s 20ms/step - loss: 0.0014 - accuracy: 1.0000\n",
      "Epoch 50/50\n",
      "1008/1008 [==============================] - ETA: 16s - loss: 5.0505e-04 - accuracy: 1.000 - ETA: 17s - loss: 4.7876e-04 - accuracy: 1.000 - ETA: 17s - loss: 7.0552e-04 - accuracy: 1.000 - ETA: 20s - loss: 6.9307e-04 - accuracy: 1.000 - ETA: 18s - loss: 8.4382e-04 - accuracy: 1.000 - ETA: 17s - loss: 8.8621e-04 - accuracy: 1.000 - ETA: 16s - loss: 0.0012 - accuracy: 1.0000    - ETA: 15s - loss: 0.0014 - accuracy: 1.000 - ETA: 14s - loss: 0.0014 - accuracy: 1.000 - ETA: 13s - loss: 0.0013 - accuracy: 1.000 - ETA: 12s - loss: 0.0013 - accuracy: 1.000 - ETA: 11s - loss: 0.0014 - accuracy: 1.000 - ETA: 11s - loss: 0.0014 - accuracy: 1.000 - ETA: 11s - loss: 0.0014 - accuracy: 1.000 - ETA: 10s - loss: 0.0014 - accuracy: 1.000 - ETA: 9s - loss: 0.0013 - accuracy: 1.000 - ETA: 9s - loss: 0.0013 - accuracy: 1.00 - ETA: 8s - loss: 0.0013 - accuracy: 1.00 - ETA: 7s - loss: 0.0014 - accuracy: 1.00 - ETA: 7s - loss: 0.0013 - accuracy: 1.00 - ETA: 6s - loss: 0.0013 - accuracy: 1.00 - ETA: 5s - loss: 0.0013 - accuracy: 1.00 - ETA: 5s - loss: 0.0014 - accuracy: 1.00 - ETA: 4s - loss: 0.0013 - accuracy: 1.00 - ETA: 4s - loss: 0.0013 - accuracy: 1.00 - ETA: 3s - loss: 0.0013 - accuracy: 1.00 - ETA: 2s - loss: 0.0013 - accuracy: 1.00 - ETA: 2s - loss: 0.0013 - accuracy: 1.00 - ETA: 1s - loss: 0.0013 - accuracy: 1.00 - ETA: 0s - loss: 0.0012 - accuracy: 1.00 - ETA: 0s - loss: 0.0012 - accuracy: 1.00 - 20s 19ms/step - loss: 0.0014 - accuracy: 1.0000\n"
     ]
    }
   ],
   "source": [
    "sam = model.fit(X_train, Y_train, batch_size=32, epochs=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(sam.history['accuracy'], \"g--\", label=\"Accuracy of training data\")\n",
    "plt.plot(sam.history['loss'], \"r--\", label=\"Loss of training data\")\n",
    "plt.title('Model Accuracy and Loss')\n",
    "plt.ylabel('Accuracy and Loss')\n",
    "plt.xlabel('Training Epoch')\n",
    "plt.ylim(0)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "432/432 [==============================] - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - ETA:  - 2s 4ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.438615341981252, 0.858134925365448]"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_test, Y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Testing Accuracy of 85.81 %"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}