tutorial_deep_learning_basics_MITmod.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "tutorial_deep_learning_basics_MITmod.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "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": "3.6.7"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DRDY0S9L-n6D"
      },
      "source": [
        "**Modified version of the MIT example for MNIST CNN**\n",
        "\n",
        "Mostly adjusted some errors with tensorflow updates.\n",
        "\n",
        "Ricardo Dalagnol (ricds@hotmail.com)<BR><BR>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fesFRiKt7EGj"
      },
      "source": [
        "![MIT Deep Learning](https://deeplearning.mit.edu/files/images/github/mit_deep_learning.png)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-hipbd137EGn"
      },
      "source": [
        "<table align=\"center\">\n",
        "  <td align=\"center\"><a target=\"_blank\" href=\"https://deeplearning.mit.edu\">\n",
        "        <img src=\"https://deeplearning.mit.edu/files/images/github/icon_mit.png\" style=\"padding-bottom:5px;\" />\n",
        "      Visit MIT Deep Learning</a></td>\n",
        "  <td align=\"center\"><a target=\"_blank\" href=\"http://colab.research.google.com/github/lexfridman/mit-deep-learning/blob/master/tutorial_deep_learning_basics/deep_learning_basics.ipynb\">\n",
        "        <img src=\"https://deeplearning.mit.edu/files/images/github/icon_google_colab.png\" style=\"padding-bottom:5px;\" />Run in Google Colab</a></td>\n",
        "  <td align=\"center\"><a target=\"_blank\" href=\"https://github.com/lexfridman/mit-deep-learning/blob/master/tutorial_deep_learning_basics/deep_learning_basics.ipynb\">\n",
        "        <img src=\"https://deeplearning.mit.edu/files/images/github/icon_github.png\" style=\"padding-bottom:5px;\"  />View Source on GitHub</a></td>\n",
        "  <td align=\"center\"><a target=\"_blank\" align=\"center\" href=\"https://www.youtube.com/watch?v=O5xeyoRL95U&list=PLrAXtmErZgOeiKm4sgNOknGvNjby9efdf\">\n",
        "        <img src=\"https://deeplearning.mit.edu/files/images/github/icon_youtube.png\" style=\"padding-bottom:5px;\" />Watch YouTube Videos</a></td>\n",
        "<!--   <td><a target=\"_blank\" href=\"link\">\n",
        "        <img src=\"image\" />text</a></td> -->\n",
        "</table>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2IkBvpMo7EGo"
      },
      "source": [
        "# Deep Learning Basics\n",
        "\n",
        "This tutorial accompanies the [lecture on Deep Learning Basics](https://www.youtube.com/watch?list=PLrAXtmErZgOeiKm4sgNOknGvNjby9efdf&v=O5xeyoRL95U) given as part of [MIT Deep Learning](https://deeplearning.mit.edu). Acknowledgement to amazing people involved is provided throughout the tutorial and at the end. You can watch the video on YouTube:\n",
        "\n",
        "[![Deep Learning Basics](https://i.imgur.com/FfQVV8q.png)](https://www.youtube.com/watch?list=PLrAXtmErZgOeiKm4sgNOknGvNjby9efdf&v=O5xeyoRL95U)\n",
        "\n",
        "In this tutorial, we mention seven important types/concepts/approaches in deep learning, introducing the first 2 and providing pointers to tutorials on the others. Here is a visual representation of the seven:\n",
        "\n",
        "![Deep learning concepts](https://i.imgur.com/EAl47rp.png)\n",
        "\n",
        "At a high-level, neural networks are either encoders, decoders, or a combination of both. Encoders find patterns in raw data to form compact, useful representations. Decoders generate new data or high-resolution useful infomation from those representations. As the lecture describes, deep learning discovers ways to **represent** the world so that we can reason about it. The rest is clever methods that help use deal effectively with visual information, language, sound (#1-6) and even act in a world based on this information and occasional rewards (#7).\n",
        "\n",
        "1. **Feed Forward Neural Networks (FFNNs)** - classification and regression based on features. See [Part 1](#Part-1:-Boston-Housing-Price-Prediction-with-Feed-Forward-Neural-Networks) of this tutorial for an example.\n",
        "2. **Convolutional Neural Networks (CNNs)** - image classification, object detection, video action recognition, etc. See [Part 2](#Part-2:-Classification-of-MNIST-Dreams-with-Convolution-Neural-Networks) of this tutorial for an example.\n",
        "3. **Recurrent Neural Networks (RNNs)** - language modeling, speech recognition/generation, etc. See [this TF tutorial on text generation](https://www.tensorflow.org/tutorials/sequences/text_generation) for an example.\n",
        "4. **Encoder Decoder Architectures** - semantic segmentation, machine translation, etc. See [our tutorial on semantic segmentation](https://github.com/lexfridman/mit-deep-learning/blob/master/tutorial_driving_scene_segmentation/tutorial_driving_scene_segmentation.ipynb) for an example.\n",
        "5. **Autoencoder** - unsupervised embeddings, denoising, etc.\n",
        "6. **Generative Adversarial Networks (GANs)** - unsupervised generation of realistic images, etc. See [this TF tutorial on DCGANs](https://github.com/tensorflow/tensorflow/blob/r1.11/tensorflow/contrib/eager/python/examples/generative_examples/dcgan.ipynb) for an example.\n",
        "7. **Deep Reinforcement Learning** - game playing, robotics in simulation, self-play, neural arhitecture search, etc. We'll be releasing notebooks on this soon and will link them here.\n",
        "\n",
        "There are selective omissions and simplifications throughout these tutorials, hopefully without losing the essence of the underlying ideas. See Einstein quote..."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "37majzDk7EGp"
      },
      "source": [
        "## Part 0: Prerequisites:\n",
        "\n",
        "We recommend that you run this this notebook in the cloud on Google Colab (see link with icon at the top) if you're not already doing so. It's the simplest way to get started. You can also [install TensorFlow locally](https://www.tensorflow.org/install/). But, again, simple is best (with caveats):\n",
        "\n",
        "![Einstein](https://i.imgur.com/vfPDHGN.png)\n",
        "\n",
        "[tf.keras](https://www.tensorflow.org/guide/keras) is the simplest way to build and train neural network models in TensorFlow. So, that's what we'll stick with in this tutorial, unless the models neccessitate a lower-level API.\n",
        "\n",
        "Note that there's [tf.keras](https://www.tensorflow.org/guide/keras) (comes with TensorFlow) and there's [Keras](https://keras.io/) (standalone). You should be using [tf.keras](https://www.tensorflow.org/guide/keras) because (1) it comes with TensorFlow so you don't need to install anything extra and (2) it comes with powerful TensorFlow-specific features."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wYrinZKs7EGq",
        "outputId": "fad7d6aa-110b-4492-86bd-b104c0cb6332"
      },
      "source": [
        "# TensorFlow and tf.keras\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dropout, Flatten, Dense\n",
        "\n",
        "# Commonly used modules\n",
        "import numpy as np\n",
        "import os\n",
        "import sys\n",
        "\n",
        "# Images, plots, display, and visualization\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import cv2\n",
        "import IPython\n",
        "from six.moves import urllib\n",
        "\n",
        "print(tf.__version__)"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.6.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "19KOxPwT7EGs"
      },
      "source": [
        "## Part 1: Boston Housing Price Prediction with Feed Forward Neural Networks\n",
        "\n",
        "Let's start with using a fully-connected neural network to do predict housing prices. The following image highlights the difference between regression and classification (see part 2). Given an observation as input, **regression** outputs a continuous value (e.g., exact temperature) and classificaiton outputs a class/category that the observation belongs to.\n",
        "\n",
        "<img src=\"https://i.imgur.com/vvSoAzg.jpg\" alt=\"classification_regression\" width=\"400\"/>\n",
        "\n",
        "For the Boston housing dataset, we get 506 rows of data, with 13 features in each. Our task is to build a regression model that takes these 13 features as input and output a single value prediction of the \"median value of owner-occupied homes (in $1000).\"\n",
        "\n",
        "Now, we load the dataset. Loading the dataset returns four NumPy arrays:\n",
        "\n",
        "* The `train_images` and `train_labels` arrays are the *training set*—the data the model uses to learn.\n",
        "* The model is tested against the *test set*, the `test_images`, and `test_labels` arrays."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S-4NsuBS7EGs",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "504f86fe-2f3a-455b-e95c-b15a47d04cb0"
      },
      "source": [
        "(train_features, train_labels), (test_features, test_labels) = keras.datasets.boston_housing.load_data()\n",
        "\n",
        "# get per-feature statistics (mean, standard deviation) from the training set to normalize by\n",
        "train_mean = np.mean(train_features, axis=0)\n",
        "train_std = np.std(train_features, axis=0)\n",
        "train_features = (train_features - train_mean) / train_std"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/boston_housing.npz\n",
            "57344/57026 [==============================] - 0s 0us/step\n",
            "65536/57026 [==================================] - 0s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7Zx0pYoh7EGt"
      },
      "source": [
        "### Build the model\n",
        "\n",
        "Building the neural network requires configuring the layers of the model, then compiling the model. First we stack a few layers together using `keras.Sequential`. Next we configure the loss function, optimizer, and metrics to monitor. These are added during the model's compile step:\n",
        "\n",
        "* *Loss function* - measures how accurate the model is during training, we want to minimize this with the optimizer.\n",
        "* *Optimizer* - how the model is updated based on the data it sees and its loss function.\n",
        "* *Metrics* - used to monitor the training and testing steps.\n",
        "\n",
        "Let's build a network with 1 hidden layer of 20 neurons, and use mean squared error (MSE) as the loss function (most common one for regression problems):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d3BV8ZZc7EGu"
      },
      "source": [
        "def build_model():\n",
        "    model = keras.Sequential([\n",
        "        Dense(20, activation=tf.nn.relu, input_shape=[len(train_features[0])]),\n",
        "        Dense(1)\n",
        "    ])\n",
        "\n",
        "    model.compile(optimizer=tf.train.AdamOptimizer(), \n",
        "                  loss='mse',\n",
        "                  metrics=['mae', 'mse'])\n",
        "    return model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "67_2OKpO7EGu"
      },
      "source": [
        "### Train the model\n",
        "\n",
        "Training the neural network model requires the following steps:\n",
        "\n",
        "1. Feed the training data to the model—in this example, the `train_features` and `train_labels` arrays.\n",
        "2. The model learns to associate features and labels.\n",
        "3. We ask the model to make predictions about a test set—in this example, the `test_features` array. We verify that the predictions match the labels from the `test_labels` array. \n",
        "\n",
        "To start training,  call the `model.fit` method—the model is \"fit\" to the training data:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tTAf-dSI7EGv",
        "outputId": "f64f27f3-8601-493b-e73c-1dbc360c0086"
      },
      "source": [
        "# this helps makes our output less verbose but still shows progress\n",
        "class PrintDot(keras.callbacks.Callback):\n",
        "    def on_epoch_end(self, epoch, logs):\n",
        "        if epoch % 100 == 0: print('')\n",
        "        print('.', end='')\n",
        "\n",
        "model = build_model()\n",
        "\n",
        "early_stop = keras.callbacks.EarlyStopping(monitor='val_loss', patience=50)\n",
        "history = model.fit(train_features, train_labels, epochs=1000, verbose=0, validation_split = 0.1,\n",
        "                    callbacks=[early_stop, PrintDot()])\n",
        "\n",
        "hist = pd.DataFrame(history.history)\n",
        "hist['epoch'] = history.epoch\n",
        "\n",
        "# show RMSE measure to compare to Kaggle leaderboard on https://www.kaggle.com/c/boston-housing/leaderboard\n",
        "rmse_final = np.sqrt(float(hist['val_mean_squared_error'].tail(1)))\n",
        "print()\n",
        "print('Final Root Mean Square Error on validation set: {}'.format(round(rmse_final, 3)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            "....................................................................................................\n",
            ".............................................................................................\n",
            "Final Root Mean Square Error on validation set: 2.359\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oyvbwdAL7EGv"
      },
      "source": [
        "Now, let's plot the loss function measure on the training and validation sets. The validation set is used to prevent overfitting ([learn more about it here](https://www.tensorflow.org/tutorials/keras/overfit_and_underfit)). However, because our network is small, the training convergence without noticeably overfitting the data as the plot shows."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "id": "WZokyMPS7EGw",
        "outputId": "9e52eeec-6271-4b8e-cfb4-1c49f9dbf640"
      },
      "source": [
        "def plot_history():\n",
        "    plt.figure()\n",
        "    plt.xlabel('Epoch')\n",
        "    plt.ylabel('Mean Square Error [Thousand Dollars$^2$]')\n",
        "    plt.plot(hist['epoch'], hist['mean_squared_error'], label='Train Error')\n",
        "    plt.plot(hist['epoch'], hist['val_mean_squared_error'], label = 'Val Error')\n",
        "    plt.legend()\n",
        "    plt.ylim([0,50])\n",
        "\n",
        "plot_history()"
      ],
      "execution_count": null,
      "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"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BPTtFuZR7EGw"
      },
      "source": [
        "Next, compare how the model performs on the test dataset:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "40UqJDEe7EGw",
        "outputId": "18e7f8e1-4cb2-4975-d8eb-b49c53dd1842"
      },
      "source": [
        "test_features_norm = (test_features - train_mean) / train_std\n",
        "mse, _, _ = model.evaluate(test_features_norm, test_labels)\n",
        "rmse = np.sqrt(mse)\n",
        "print('Root Mean Square Error on test set: {}'.format(round(rmse, 3)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "102/102 [==============================] - 0s 44us/step\n",
            "Root Mean Square Error on test set: 4.244\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OC7rRnf37EGx"
      },
      "source": [
        "Compare the RMSE measure you get to the [Kaggle leaderboard](https://www.kaggle.com/c/boston-housing/leaderboard). An RMSE of 2.651 puts us in 5th place."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kADo2smS7EGx"
      },
      "source": [
        "## Part 2: Classification of MNIST Dreams with Convolutional Neural Networks\n",
        "\n",
        "Next, let's build a convolutional neural network (CNN) classifier to classify images of handwritten digits in the MNIST dataset with a twist where we test our classifier on high-resolution hand-written digits from outside the dataset."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PjZpDdWH7EGx"
      },
      "source": [
        "# Set common constants\n",
        "this_repo_url = 'https://github.com/lexfridman/mit-deep-learning/raw/master/'\n",
        "this_tutorial_url = this_repo_url + 'tutorial_deep_learning_basics'"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BWeSgpAl7EGx"
      },
      "source": [
        "The MNIST dataset containss 70,000 grayscale images of handwritten digits at a resolution of 28 by 28 pixels. The task is to take one of these images as input and predict the most likely digit contained in the image (along with a relative confidence in this prediction):\n",
        "\n",
        "<img src=\"https://i.imgur.com/ITrm9x4.png\" width=\"500px\">\n",
        "\n",
        "Now, we load the dataset. The images are 28x28 NumPy arrays, with pixel values ranging between 0 and 255. The *labels* are an array of integers, ranging from 0 to 9."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EgrCyQE07EGy",
        "outputId": "9b4d3e32-cde2-4d18-8d8a-4a50aeae63ef"
      },
      "source": [
        "(train_images, train_labels), (test_images, test_labels) = keras.datasets.mnist.load_data()\n",
        "\n",
        "# reshape images to specify that it's a single channel\n",
        "train_images = train_images.reshape(train_images.shape[0], 28, 28, 1)\n",
        "test_images = test_images.reshape(test_images.shape[0], 28, 28, 1)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n",
            "11501568/11490434 [==============================] - 0s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zb9W9PuF7EGy"
      },
      "source": [
        "We scale these values to a range of 0 to 1 before feeding to the neural network model. For this, we divide the values by 255. It's important that the *training set* and the *testing set* are preprocessed in the same way:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zGtZtri77EGy"
      },
      "source": [
        "def preprocess_images(imgs): # should work for both a single image and multiple images\n",
        "    sample_img = imgs if len(imgs.shape) == 2 else imgs[0]\n",
        "    assert sample_img.shape in [(28, 28, 1), (28, 28)], sample_img.shape # make sure images are 28x28 and single-channel (grayscale)\n",
        "    return imgs / 255.0\n",
        "\n",
        "train_images = preprocess_images(train_images)\n",
        "test_images = preprocess_images(test_images)"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nM0mrkW-7EGy"
      },
      "source": [
        "Display the first 5 images from the *training set* and display the class name below each image. Verify that the data is in the correct format and we're ready to build and train the network."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 141
        },
        "id": "WtwR4ELh7EGy",
        "outputId": "5d01904b-2791-4a15-8acc-5201aceb8512"
      },
      "source": [
        "plt.figure(figsize=(10,2))\n",
        "for i in range(5):\n",
        "    plt.subplot(1,5,i+1)\n",
        "    plt.xticks([])\n",
        "    plt.yticks([])\n",
        "    plt.grid(False)\n",
        "    plt.imshow(train_images[i].reshape(28, 28), cmap=plt.cm.binary)\n",
        "    plt.xlabel(train_labels[i])"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x144 with 5 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fhNjx9K47EGz"
      },
      "source": [
        "### Build the model\n",
        "\n",
        "Building the neural network requires configuring the layers of the model, then compiling the model. In many cases, this can be reduced to simply stacking together layers:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FfqphmPY7EGz"
      },
      "source": [
        "model = keras.Sequential()\n",
        "# 32 convolution filters used each of size 3x3\n",
        "model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(28, 28, 1)))\n",
        "# 64 convolution filters used each of size 3x3\n",
        "model.add(Conv2D(64, (3, 3), activation='relu'))\n",
        "# choose the best features via pooling\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "# randomly turn neurons on and off to improve convergence\n",
        "model.add(Dropout(0.25))\n",
        "# flatten since too many dimensions, we only want a classification output\n",
        "model.add(Flatten())\n",
        "# fully connected to get all relevant data\n",
        "model.add(Dense(128, activation='relu'))\n",
        "# one more dropout\n",
        "model.add(Dropout(0.5))\n",
        "# output a softmax to squash the matrix into output probabilities\n",
        "model.add(Dense(10, activation='softmax'))"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JQPlFRmn-Y_k"
      },
      "source": [
        "Look at the model summary"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n0bnUUQA-Xvf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fc34f466-8c42-43a7-ced9-518f1453aa0f"
      },
      "source": [
        "# summary\n",
        "model.summary()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 26, 26, 32)        320       \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 24, 24, 64)        18496     \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 12, 12, 64)        0         \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 12, 12, 64)        0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 9216)              0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 128)               1179776   \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 10)                1290      \n",
            "=================================================================\n",
            "Total params: 1,199,882\n",
            "Trainable params: 1,199,882\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pwg6C-fc7EGz"
      },
      "source": [
        "Before the model is ready for training, it needs a few more settings. These are added during the model's *compile* step:\n",
        "\n",
        "* *Loss function* - measures how accurate the model is during training, we want to minimize this with the optimizer.\n",
        "* *Optimizer* - how the model is updated based on the data it sees and its loss function.\n",
        "* *Metrics* - used to monitor the training and testing steps. \"accuracy\" is the fraction of images that are correctly classified."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-ZKJ8Twu7EGz"
      },
      "source": [
        "model.compile(optimizer=tf.optimizers.Adam(), \n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3at-_pBW7EG0"
      },
      "source": [
        "### Train the model\n",
        "\n",
        "Training the neural network model requires the following steps:\n",
        "\n",
        "1. Feed the training data to the model—in this example, the `train_images` and `train_labels` arrays.\n",
        "2. The model learns to associate images and labels.\n",
        "3. We ask the model to make predictions about a test set—in this example, the `test_images` array. We verify that the predictions match the labels from the `test_labels` array. \n",
        "\n",
        "To start training,  call the `model.fit` method—the model is \"fit\" to the training data:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EfYW1kCU7EG0",
        "outputId": "d303b1ce-67e6-41da-8f53-be594a525551"
      },
      "source": [
        "import math\n",
        "history = model.fit(train_images, train_labels, epochs=5, steps_per_epoch = math.floor(60000/32))"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "2187/2187 [==============================] - 45s 7ms/step - loss: 0.2028 - accuracy: 0.9387\n",
            "Epoch 2/5\n",
            "2187/2187 [==============================] - 14s 7ms/step - loss: 0.0858 - accuracy: 0.9745\n",
            "Epoch 3/5\n",
            "2187/2187 [==============================] - 15s 7ms/step - loss: 0.0641 - accuracy: 0.9803\n",
            "Epoch 4/5\n",
            "2187/2187 [==============================] - 15s 7ms/step - loss: 0.0536 - accuracy: 0.9836\n",
            "Epoch 5/5\n",
            "1964/2187 [=========================>....] - ETA: 1s - loss: 0.0423 - accuracy: 0.9866WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 10935 batches). You may need to use the repeat() function when building your dataset.\n",
            "2187/2187 [==============================] - 13s 6ms/step - loss: 0.0422 - accuracy: 0.9866\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cPRJH97K7EG0"
      },
      "source": [
        "As the model trains, the loss and accuracy metrics are displayed. This model reaches an accuracy of about 98.68% on the training data."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "16cbD8No7EG0"
      },
      "source": [
        "### Evaluate accuracy\n",
        "\n",
        "Next, compare how the model performs on the test dataset:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IXfZq_eN7EG0",
        "outputId": "fe7a4e72-51da-42b4-e755-fa2c20aa06e4"
      },
      "source": [
        "print(test_images.shape)\n",
        "test_loss, test_acc = model.evaluate(test_images, test_labels)\n",
        "\n",
        "print('Test accuracy:', test_acc)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(10000, 28, 28, 1)\n",
            "313/313 [==============================] - 2s 4ms/step - loss: 0.0275 - accuracy: 0.9913\n",
            "Test accuracy: 0.9912999868392944\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kvY7TJo07EG0"
      },
      "source": [
        "Often times, the accuracy on the test dataset is a little less than the accuracy on the training dataset. This gap between training accuracy and test accuracy is an example of *overfitting*. In our case, the accuracy is better at 99.19%! This is, in part, due to successful regularization accomplished with the Dropout layers."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2QYDvXtTQrv6"
      },
      "source": [
        "### Make predictions to test dataset\n",
        "\n",
        "With the model trained, we used it to predict some of the test images - not used in training."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 141
        },
        "id": "cTCtH7kbMYvX",
        "outputId": "4b148d99-791e-45ec-f873-6def12706fa2"
      },
      "source": [
        "# predict the class\n",
        "preds = model.predict(test_images)\n",
        "guess = np.argmax(preds, 1)\n",
        "perc = np.rint(preds * 100).astype(int)\n",
        "\n",
        "# five random numbers\n",
        "from numpy.random import randint\n",
        "values = randint(0, 10000, 5)\n",
        "\n",
        "# plot\n",
        "plt.figure(figsize=(10,2))\n",
        "for i in range(5):\n",
        "    plt.subplot(1,5,i+1)\n",
        "    plt.xticks([])\n",
        "    plt.yticks([])\n",
        "    plt.grid(False)\n",
        "    plt.imshow(test_images[values[i]].reshape(28, 28), cmap=plt.cm.binary)\n",
        "    plt.xlabel(guess[values[i]])"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x144 with 5 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1UeXJ9K_7EG1"
      },
      "source": [
        "### Make predictions to MNIST dream dataset\n",
        "\n",
        "With the model trained, we can use it to make predictions about some images. Let's step outside the MNIST dataset for that and go with the beautiful high-resolution images generated by a mixture of CPPN, GAN, VAE. See [great blog post by hardmaru](http://blog.otoro.net/2016/04/01/generating-large-images-from-latent-vectors/) for the source data and a description of how these morphed animations are generated:\n",
        "\n",
        "![MNIST dream](https://i.imgur.com/OrUJs9V.gif)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 377
        },
        "id": "_9BV1_Ox7EG1",
        "outputId": "73a28611-c268-447d-8a9f-73a9606fb5d5"
      },
      "source": [
        "mnist_dream_path = 'images/mnist_dream.mp4'\n",
        "mnist_prediction_path = 'images/mnist_dream_predicted.mp4'\n",
        "\n",
        "# download the video if running in Colab\n",
        "if not os.path.isfile(mnist_dream_path): \n",
        "    print('downloading the sample video...')\n",
        "    vid_url = this_tutorial_url + '/' + mnist_dream_path\n",
        "    \n",
        "    mnist_dream_path = urllib.request.urlretrieve(vid_url)[0]\n",
        "                                                                                                  \n",
        "def cv2_imshow(img):\n",
        "    ret = cv2.imencode('.png', img)[1].tobytes() \n",
        "    img_ip = IPython.display.Image(data=ret)\n",
        "    IPython.display.display(img_ip)\n",
        "\n",
        "cap = cv2.VideoCapture(mnist_dream_path) \n",
        "vw = None\n",
        "frame = -1 # counter for debugging (mostly), 0-indexed\n",
        "\n",
        "# go through all the frames and run our classifier on the high res MNIST images as they morph from number to number\n",
        "while True: # should 481 frames\n",
        "    frame += 1\n",
        "    ret, img = cap.read()\n",
        "    if not ret: break\n",
        "               \n",
        "    assert img.shape[0] == img.shape[1] # should be a square\n",
        "    if img.shape[0] != 720:\n",
        "        img = cv2.resize(img, (720, 720))\n",
        "       \n",
        "    #preprocess the image for prediction\n",
        "    img_proc = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
        "    img_proc = cv2.resize(img_proc, (28, 28))\n",
        "    img_proc = preprocess_images(img_proc)\n",
        "    img_proc = 1 - img_proc # inverse since training dataset is white text with black background\n",
        "\n",
        "    net_in = np.expand_dims(img_proc, axis=0) # expand dimension to specify batch size of 1\n",
        "    net_in = np.expand_dims(net_in, axis=3) # expand dimension to specify number of channels\n",
        "    \n",
        "    preds = model.predict(net_in)[0]\n",
        "    guess = np.argmax(preds)\n",
        "    perc = np.rint(preds * 100).astype(int)\n",
        "    \n",
        "    img = 255 - img\n",
        "    pad_color = 0\n",
        "    img = np.pad(img, ((0,0), (0,1280-720), (0,0)), mode='constant', constant_values=(pad_color))  \n",
        "    \n",
        "    line_type = cv2.LINE_AA\n",
        "    font_face = cv2.FONT_HERSHEY_SIMPLEX\n",
        "    font_scale = 1.3        \n",
        "    thickness = 2\n",
        "    x, y = 740, 60\n",
        "    color = (255, 255, 255)\n",
        "    \n",
        "    text = \"Neural Network Output:\"\n",
        "    cv2.putText(img, text=text, org=(x, y), fontScale=font_scale, fontFace=font_face, thickness=thickness,\n",
        "                    color=color, lineType=line_type)\n",
        "    \n",
        "    text = \"Input:\"\n",
        "    cv2.putText(img, text=text, org=(30, y), fontScale=font_scale, fontFace=font_face, thickness=thickness,\n",
        "                    color=color, lineType=line_type)   \n",
        "        \n",
        "    y = 130\n",
        "    for i, p in enumerate(perc):\n",
        "        if i == guess: color = (255, 218, 158)\n",
        "        else: color = (100, 100, 100)\n",
        "            \n",
        "        rect_width = 0\n",
        "        if p > 0: rect_width = int(p * 3.3)\n",
        "        \n",
        "        rect_start = 180\n",
        "        cv2.rectangle(img, (x+rect_start, y-5), (x+rect_start+rect_width, y-20), color, -1)\n",
        "\n",
        "        text = '{}: {:>3}%'.format(i, int(p))\n",
        "        cv2.putText(img, text=text, org=(x, y), fontScale=font_scale, fontFace=font_face, thickness=thickness,\n",
        "                    color=color, lineType=line_type)\n",
        "        y += 60\n",
        "    \n",
        "    # if you don't want to save the output as a video, set this to False\n",
        "    save_video = True\n",
        "    \n",
        "    if save_video:\n",
        "        if vw is None:\n",
        "            codec = cv2.VideoWriter_fourcc(*'DIVX')\n",
        "            vid_width_height = img.shape[1], img.shape[0]\n",
        "            vw = cv2.VideoWriter(mnist_prediction_path, codec, 30, vid_width_height)\n",
        "        # 15 fps above doesn't work robustly so we right frame twice at 30 fps\n",
        "        vw.write(img)\n",
        "        vw.write(img)\n",
        "    \n",
        "    # scale down image for display\n",
        "    img_disp = cv2.resize(img, (0,0), fx=0.5, fy=0.5)\n",
        "    cv2_imshow(img_disp)\n",
        "    IPython.display.clear_output(wait=True)\n",
        "        \n",
        "cap.release()\n",
        "if vw is not None:\n",
        "    vw.release()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rOsZ-6oH7EG1"
      },
      "source": [
        "The above shows the prediction of the network by choosing the neuron with the highest output. While the output layer values add 1 to one, these do not reflect well-calibrated measures of \"uncertainty\". Often, the network is overly confident about the top choice that does not reflect a learned measure of probability. If everything ran correctly you should get an animation like this:\n",
        "\n",
        "![MNIST dream predictions](https://i.imgur.com/eMF9FOG.gif)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PFvHv9Hn7EG1"
      },
      "source": [
        "## Acknowledgements\n",
        "\n",
        "The contents of this tutorial is based on and inspired by the work of [TensorFlow team](https://www.tensorflow.org) (see their [Colab notebooks](https://www.tensorflow.org/tutorials/)), our [MIT Human-Centered AI team](https://hcai.mit.edu), and individual pieces referenced in the [MIT Deep Learning](https://deeplearning.mit.edu) course slides."
      ]
    }
  ]
}