13_cnn_mnist.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "13_cnn_mnist.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DpII-qgno6EX",
        "colab_type": "text"
      },
      "source": [
        "# Convolutional Neural Network - MNIST\n",
        "\n",
        "Make Your First GAN With PyTorch, 2020"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B6g6LPI9Xjgl",
        "colab_type": "code",
        "outputId": "b3881bd7-e303-4acf-cfcb-1580dde34f12",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "# mount Drive to access data files\n",
        "\n",
        "from google.colab import drive\n",
        "drive.mount('./mount')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at ./mount; to attempt to forcibly remount, call drive.mount(\"./mount\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7bkrfSHHAPUB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# import libraries\n",
        "\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from torch.utils.data import Dataset\n",
        "\n",
        "import pandas\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "kZpq_7wQPNJD"
      },
      "source": [
        "# Helper Functions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eAcDGdnNPVOo",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# modified from https://github.com/pytorch/vision/issues/720\n",
        "\n",
        "class View(nn.Module):\n",
        "    def __init__(self, shape):\n",
        "        super().__init__()\n",
        "        self.shape = shape,\n",
        "\n",
        "    def forward(self, x):\n",
        "        return x.view(*self.shape)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o91c-ALVmw0n",
        "colab_type": "text"
      },
      "source": [
        "## DataSet Class"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VImRycohm3Ct",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# dataset class\n",
        "\n",
        "class MnistDataset(Dataset):\n",
        "    \n",
        "    def __init__(self, csv_file):\n",
        "        self.data_df = pandas.read_csv(csv_file, header=None)\n",
        "        pass\n",
        "    \n",
        "    def __len__(self):\n",
        "        return len(self.data_df)\n",
        "    \n",
        "    def __getitem__(self, index):\n",
        "        # image target (label)\n",
        "        label = self.data_df.iloc[index,0]\n",
        "        target = torch.zeros((10))\n",
        "        target[label] = 1.0\n",
        "        \n",
        "        # image data, normalised from 0-255 to 0-1\n",
        "        image_values = torch.FloatTensor(self.data_df.iloc[index,1:].values) / 255.0\n",
        "        \n",
        "        # return label, image data tensor and target tensor\n",
        "        return label, image_values, target\n",
        "    \n",
        "    def plot_image(self, index):\n",
        "        img = self.data_df.iloc[index,1:].values.reshape(28,28)\n",
        "        plt.title(\"label = \" + str(self.data_df.iloc[index,0]))\n",
        "        plt.imshow(img, interpolation='none', cmap='Blues')\n",
        "        pass\n",
        "    \n",
        "    pass\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AJ_E9wgYQybx",
        "colab_type": "text"
      },
      "source": [
        "## Load Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wioFGZ4mKUqj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "mnist_dataset = MnistDataset('mount/My Drive/Colab Notebooks/myo_gan/mnist_data/mnist_train.csv')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bnq2wNUhLZRg",
        "colab_type": "code",
        "outputId": "a44f8dbd-276b-46ae-f01b-fbf31cfcd419",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "# check data contains images\n",
        "\n",
        "mnist_dataset.plot_image(9)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAEICAYAAACZA4KlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R1VyK4VQpNYA",
        "colab_type": "text"
      },
      "source": [
        "## Neural Network Class"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MrgKlTsRAfSk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# classifier class\n",
        "\n",
        "class Classifier(nn.Module):\n",
        "    \n",
        "    def __init__(self):\n",
        "        # initialise parent pytorch class\n",
        "        super().__init__()\n",
        "        \n",
        "        # define neural network layers\n",
        "        self.model = nn.Sequential(\n",
        "            # expand 1 to 10 filters\n",
        "            nn.Conv2d(1, 10, kernel_size=5, stride=2),\n",
        "            nn.LeakyReLU(0.02),\n",
        "            nn.BatchNorm2d(10),\n",
        "        \n",
        "            # 10 filters to 10 filters\n",
        "            nn.Conv2d(10, 10, kernel_size=3, stride=2),\n",
        "            nn.LeakyReLU(0.02),\n",
        "            nn.BatchNorm2d(10),\n",
        "            \n",
        "            View(250),\n",
        "            nn.Linear(250, 10),\n",
        "            nn.Sigmoid()\n",
        "        )\n",
        "        \n",
        "        # create loss function\n",
        "        self.loss_function = nn.BCELoss()\n",
        "\n",
        "        # create optimiser, using simple stochastic gradient descent\n",
        "        self.optimiser = torch.optim.Adam(self.parameters())\n",
        "\n",
        "        # counter and accumulator for progress\n",
        "        self.counter = 0\n",
        "        self.progress = []\n",
        "\n",
        "        pass\n",
        "    \n",
        "    \n",
        "    def forward(self, inputs):\n",
        "        # simply run model\n",
        "        return self.model(inputs)\n",
        "    \n",
        "    \n",
        "    def train(self, inputs, targets):\n",
        "        # calculate the output of the network\n",
        "        outputs = self.forward(inputs)\n",
        "        \n",
        "        # calculate loss\n",
        "        loss = self.loss_function(outputs, targets)\n",
        "\n",
        "        # increase counter and accumulate error every 10\n",
        "        self.counter += 1\n",
        "        if (self.counter % 10 == 0):\n",
        "            self.progress.append(loss.item())\n",
        "            pass\n",
        "        if (self.counter % 10000 == 0):\n",
        "            print(\"counter = \", self.counter)\n",
        "            pass\n",
        "\n",
        "        # zero gradients, perform a backward pass, and update the weights\n",
        "        self.optimiser.zero_grad()\n",
        "        loss.backward()\n",
        "        self.optimiser.step()\n",
        "\n",
        "        pass\n",
        "    \n",
        "    \n",
        "    def plot_progress(self):\n",
        "        df = pandas.DataFrame(self.progress, columns=['loss'])\n",
        "        df.plot(ylim=(0, 1.0), figsize=(16,8), alpha=0.1, marker='.', grid=True, yticks=(0, 0.25, 0.5))\n",
        "        pass\n",
        "    \n",
        "    pass"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cwGBnd4xSLBL",
        "colab_type": "text"
      },
      "source": [
        "## Train Neural Network"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VLS3pXEevWq9",
        "colab_type": "code",
        "outputId": "e656fd83-0a6c-4517-c639-68c9d411c401",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        }
      },
      "source": [
        "%%time \n",
        "# create neural network\n",
        "\n",
        "C = Classifier()\n",
        "\n",
        "# train network on MNIST data set\n",
        "\n",
        "epochs = 3\n",
        "\n",
        "for i in range(epochs):\n",
        "    print('training epoch', i+1, \"of\", epochs)\n",
        "    for label, image_data_tensor, target_tensor in mnist_dataset:\n",
        "        C.train(image_data_tensor.view(1, 1, 28, 28), target_tensor)\n",
        "        pass\n",
        "    pass"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "training epoch 1 of 3\n",
            "counter =  10000\n",
            "counter =  20000\n",
            "counter =  30000\n",
            "counter =  40000\n",
            "counter =  50000\n",
            "counter =  60000\n",
            "training epoch 2 of 3\n",
            "counter =  70000\n",
            "counter =  80000\n",
            "counter =  90000\n",
            "counter =  100000\n",
            "counter =  110000\n",
            "counter =  120000\n",
            "training epoch 3 of 3\n",
            "counter =  130000\n",
            "counter =  140000\n",
            "counter =  150000\n",
            "counter =  160000\n",
            "counter =  170000\n",
            "counter =  180000\n",
            "CPU times: user 8min 57s, sys: 26.1 s, total: 9min 23s\n",
            "Wall time: 9min 32s\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G046xt9ySheM",
        "colab_type": "code",
        "outputId": "f62e394e-499c-4c32-b569-ce3cf602d281",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 483
        }
      },
      "source": [
        "# plot classifier error\n",
        "\n",
        "C.plot_progress()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6gAAAHSCAYAAADhZ+amAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n50Lw/wORA5xzb9Cw3AAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1152x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Sv7R988-4pCv",
        "colab_type": "text"
      },
      "source": [
        "## Classification Example"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tfi78c5XTLD3",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# load MNIST test data\n",
        "\n",
        "mnist_test_dataset = MnistDataset('mount/My Drive/Colab Notebooks/gan/mnist_data/mnist_test.csv')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cdQCvzor-QWx",
        "colab_type": "code",
        "outputId": "218a841e-c9e4-4d23-b08a-23b5613ba710",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        }
      },
      "source": [
        "# pick a record\n",
        "record = 19\n",
        "\n",
        "# plot image and correct label\n",
        "mnist_test_dataset.plot_image(record)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAEICAYAAACZA4KlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nu1kSDrtZEg67WRIOu1kS/w9XtbXIcGokPAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CqZmjoqs-iJH",
        "colab_type": "code",
        "outputId": "9e719654-5d0c-47fe-d766-25c23421ea98",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        }
      },
      "source": [
        "# visualise the answer given by the neural network\n",
        "\n",
        "image_data = mnist_test_dataset[record][1]\n",
        "\n",
        "# query from trained network\n",
        "output = C.forward(image_data.view(1,1,28,28))\n",
        "\n",
        "# plot output tensor\n",
        "pandas.DataFrame(output.detach().numpy()).plot(kind='bar', legend=False, ylim=(0,1))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f04d09aef98>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD5CAYAAAA3Os7hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0CWpEQZdkhrxv/HNZWf0zuJnAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yYobxR3rMne7",
        "colab_type": "text"
      },
      "source": [
        "## Classifier Performance"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dkcHXgffGcKn",
        "colab_type": "code",
        "outputId": "d339ae8d-d4c5-45bb-f9d3-d0b9c937dfd9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "# test trained neural network on training data\n",
        "\n",
        "score = 0\n",
        "items = 0\n",
        "\n",
        "for label, image_data_tensor, target_tensor in mnist_test_dataset:\n",
        "    answer = C.forward(image_data_tensor.view(1,1,28,28)).detach().numpy()\n",
        "    if (answer.argmax() == label):\n",
        "        score += 1\n",
        "        pass\n",
        "    items += 1\n",
        "    \n",
        "    pass\n",
        "\n",
        "print(score, items, score/items)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "9814 10000 0.9814\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jzOlSWGbVGjK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}