l08c04_time_windows.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "l08c04_time_windows.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Za8-Nr5k11fh"
      },
      "source": [
        "##### Copyright 2018 The TensorFlow Authors."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "cellView": "form",
        "id": "Eq10uEbw0E4l"
      },
      "source": [
        "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ou0PGp_4icRo"
      },
      "source": [
        "# Time windows"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "93b0GzKph0jK"
      },
      "source": [
        "<table class=\"tfo-notebook-buttons\" align=\"left\">\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/examples/blob/master/courses/udacity_intro_to_tensorflow_for_deep_learning/l08c04_time_windows.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
        "  </td>\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://github.com/tensorflow/examples/blob/master/courses/udacity_intro_to_tensorflow_for_deep_learning/l08c04_time_windows.ipynb\"><img src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" />View source on GitHub</a>\n",
        "  </td>\n",
        "</table>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vidayERjaO5q"
      },
      "source": [
        "## Setup"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gqWabzlJ63nL"
      },
      "source": [
        "import tensorflow as tf"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ViWVB9qd8OIR"
      },
      "source": [
        "## Time Windows\n",
        "\n",
        "First, we will train a model to forecast the next step given the previous 20 steps, therefore, we need to create a dataset of 20-step windows for training."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bgJkwtq88OIS",
        "outputId": "6ff51084-b595-4886-bde5-64caf37a7706",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "for val in dataset:\n",
        "    print(val.numpy())"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0\n",
            "1\n",
            "2\n",
            "3\n",
            "4\n",
            "5\n",
            "6\n",
            "7\n",
            "8\n",
            "9\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ad8C65JV8OIT",
        "outputId": "b2178a69-38dc-4970-8e87-5af2cf0358ec",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "dataset = dataset.window(5, shift=1)\n",
        "for window_dataset in dataset:\n",
        "    for val in window_dataset:\n",
        "        print(val.numpy(), end=\" \")\n",
        "    print()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0 1 2 3 4 \n",
            "1 2 3 4 5 \n",
            "2 3 4 5 6 \n",
            "3 4 5 6 7 \n",
            "4 5 6 7 8 \n",
            "5 6 7 8 9 \n",
            "6 7 8 9 \n",
            "7 8 9 \n",
            "8 9 \n",
            "9 \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AQtmODsi8OIU",
        "outputId": "fa39738f-378c-4081-f493-57eb7de55c60",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "dataset = dataset.window(5, shift=1, drop_remainder=True)\n",
        "for window_dataset in dataset:\n",
        "    for val in window_dataset:\n",
        "        print(val.numpy(), end=\" \")\n",
        "    print()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0 1 2 3 4 \n",
            "1 2 3 4 5 \n",
            "2 3 4 5 6 \n",
            "3 4 5 6 7 \n",
            "4 5 6 7 8 \n",
            "5 6 7 8 9 \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kTRHiWxi8OIW",
        "outputId": "73cdd8c7-d243-4897-a94d-8da0bb2c1118",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "dataset = dataset.window(5, shift=1, drop_remainder=True)\n",
        "dataset = dataset.flat_map(lambda window: window.batch(5))\n",
        "for window in dataset:\n",
        "    print(window.numpy())"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0 1 2 3 4]\n",
            "[1 2 3 4 5]\n",
            "[2 3 4 5 6]\n",
            "[3 4 5 6 7]\n",
            "[4 5 6 7 8]\n",
            "[5 6 7 8 9]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iPsQbWHb8OIX",
        "outputId": "b8d31d99-abe1-4311-da30-c0483e264663",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "dataset = dataset.window(5, shift=1, drop_remainder=True)\n",
        "dataset = dataset.flat_map(lambda window: window.batch(5))\n",
        "dataset = dataset.map(lambda window: (window[:-1], window[-1:]))\n",
        "for x, y in dataset:\n",
        "    print(x.numpy(), y.numpy())"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0 1 2 3] [4]\n",
            "[1 2 3 4] [5]\n",
            "[2 3 4 5] [6]\n",
            "[3 4 5 6] [7]\n",
            "[4 5 6 7] [8]\n",
            "[5 6 7 8] [9]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hzp7RD6_8OIY",
        "outputId": "5a9a9810-1a8c-4d75-bc21-b01127cb8560",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "dataset = dataset.window(5, shift=1, drop_remainder=True)\n",
        "dataset = dataset.flat_map(lambda window: window.batch(5))\n",
        "dataset = dataset.map(lambda window: (window[:-1], window[-1:]))\n",
        "dataset = dataset.shuffle(buffer_size=10)\n",
        "for x, y in dataset:\n",
        "    print(x.numpy(), y.numpy())"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[5 6 7 8] [9]\n",
            "[2 3 4 5] [6]\n",
            "[1 2 3 4] [5]\n",
            "[4 5 6 7] [8]\n",
            "[3 4 5 6] [7]\n",
            "[0 1 2 3] [4]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y70nV0EI8OIZ",
        "outputId": "ece15e01-c2d4-4ea6-de5e-5e72e3016ded",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "dataset = tf.data.Dataset.range(10)\n",
        "dataset = dataset.window(5, shift=1, drop_remainder=True)\n",
        "dataset = dataset.flat_map(lambda window: window.batch(5))\n",
        "dataset = dataset.map(lambda window: (window[:-1], window[-1:]))\n",
        "dataset = dataset.shuffle(buffer_size=10)\n",
        "dataset = dataset.batch(2).prefetch(1)\n",
        "for x, y in dataset:\n",
        "    print(\"x =\", x.numpy())\n",
        "    print(\"y =\", y.numpy())"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "x = [[1 2 3 4]\n",
            " [4 5 6 7]]\n",
            "y = [[5]\n",
            " [8]]\n",
            "x = [[5 6 7 8]\n",
            " [2 3 4 5]]\n",
            "y = [[9]\n",
            " [6]]\n",
            "x = [[3 4 5 6]\n",
            " [0 1 2 3]]\n",
            "y = [[7]\n",
            " [4]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1tl-0BOKkEtk"
      },
      "source": [
        "def window_dataset(series, window_size, batch_size=32,\n",
        "                   shuffle_buffer=1000):\n",
        "    dataset = tf.data.Dataset.from_tensor_slices(series)\n",
        "    dataset = dataset.window(window_size + 1, shift=1, drop_remainder=True)\n",
        "    dataset = dataset.flat_map(lambda window: window.batch(window_size + 1))\n",
        "    dataset = dataset.shuffle(shuffle_buffer)\n",
        "    dataset = dataset.map(lambda window: (window[:-1], window[-1]))\n",
        "    dataset = dataset.batch(batch_size).prefetch(1)\n",
        "    return dataset"
      ],
      "execution_count": 10,
      "outputs": []
    }
  ]
}