ConvNN_MNIST.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ConvNN_MNIST.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "[View in Colaboratory](https://colab.research.google.com/github/gamante91/pytorch_notebooks/blob/master/ConvNN_MNIST.ipynb)"
      ]
    },
    {
      "metadata": {
        "id": "-fy47lL9J0Zw",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "24ac0634-08a7-482a-817f-496369f66b87"
      },
      "cell_type": "code",
      "source": [
        "# http://pytorch.org/\n",
        "from os import path\n",
        "from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag\n",
        "platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())\n",
        "\n",
        "accelerator = 'cu80' #'cu80' if path.exists('/opt/bin/nvidia-smi') else 'cpu'\n",
        "print('Platform:', platform, 'Accelerator:', accelerator)\n",
        "\n",
        "!pip install --upgrade torch\n",
        "!pip install --upgrade torchvision"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Platform: cp36-cp36m Accelerator: cu80\n",
            "Requirement already up-to-date: torch in /usr/local/lib/python3.6/dist-packages (0.4.1)\n",
            "Requirement already up-to-date: torchvision in /usr/local/lib/python3.6/dist-packages (0.2.1)\n",
            "Requirement already satisfied, skipping upgrade: six in /usr/local/lib/python3.6/dist-packages (from torchvision) (1.11.0)\n",
            "Requirement already satisfied, skipping upgrade: torch in /usr/local/lib/python3.6/dist-packages (from torchvision) (0.4.1)\n",
            "Requirement already satisfied, skipping upgrade: numpy in /usr/local/lib/python3.6/dist-packages (from torchvision) (1.14.5)\n",
            "Requirement already satisfied, skipping upgrade: pillow>=4.1.1 in /usr/local/lib/python3.6/dist-packages (from torchvision) (5.2.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "voFivxs_KWd4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "8fc90bfc-3690-4901-d4a0-88393d319180"
      },
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Device configuration\n",
        "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
        "\n",
        "print('Torch', torch.__version__, 'CUDA', torch.version.cuda)\n",
        "print('Device:', device)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Torch 0.4.1 CUDA 9.0.176\n",
            "Device: cuda:0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Zd-UXUK5KYu3",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Hyper-parameter\n",
        "batch_size = 100\n",
        "\n",
        "# Load the MNIST dataset with torchvision \n",
        "train_dataset = torchvision.datasets.MNIST(root='../../data', \n",
        "                                           train=True, \n",
        "                                           transform=transforms.ToTensor(),  \n",
        "                                           download=True)\n",
        "\n",
        "test_dataset = torchvision.datasets.MNIST(root='../../data', \n",
        "                                          train=False, \n",
        "                                          transform=transforms.ToTensor())\n",
        "\n",
        "# Data loader\n",
        "train_loader = torch.utils.data.DataLoader(dataset=train_dataset, \n",
        "                                           batch_size=batch_size, \n",
        "                                           shuffle=True)\n",
        "\n",
        "test_loader = torch.utils.data.DataLoader(dataset=test_dataset, \n",
        "                                          batch_size=batch_size, \n",
        "                                          shuffle=False)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "QOXYZCNaMONI",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Hyper-parameters\n",
        "num_epochs = 20\n",
        "num_classes = 10\n",
        "batch_size = 100\n",
        "learning_rate = 0.001"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "i7LXTfM9KgEa",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Define Convolutional Neural Network model\n",
        "class ConvNet(nn.Module):\n",
        "    def __init__(self, num_classes=10):\n",
        "        super(ConvNet, self).__init__()\n",
        "        self.layer1 = nn.Sequential(\n",
        "            nn.Conv2d(1, 16, kernel_size=5, stride=1, padding=2),\n",
        "            nn.BatchNorm2d(16),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2, stride=2))\n",
        "        self.layer2 = nn.Sequential(\n",
        "            nn.Conv2d(16, 32, kernel_size=5, stride=1, padding=2),\n",
        "            nn.BatchNorm2d(32),\n",
        "            nn.ReLU(),\n",
        "            nn.MaxPool2d(kernel_size=2, stride=2))\n",
        "        self.fc = nn.Linear(7*7*32, num_classes)\n",
        "        \n",
        "    def forward(self, x):\n",
        "        out = self.layer1(x)\n",
        "        out = self.layer2(out)\n",
        "        out = out.reshape(out.size(0), -1)\n",
        "        out = self.fc(out)\n",
        "        return out"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "u4az4IfXK8Fj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 289
        },
        "outputId": "9129d69f-0992-40cb-855a-4e8721cf19a8"
      },
      "cell_type": "code",
      "source": [
        "# Instanciate the model\n",
        "model = ConvNet(num_classes).to(device)\n",
        "\n",
        "print(\"Model description:\\n\", model)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model description:\n",
            " ConvNet(\n",
            "  (layer1): Sequential(\n",
            "    (0): Conv2d(1, 16, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
            "    (1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    (2): ReLU()\n",
            "    (3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
            "  )\n",
            "  (layer2): Sequential(\n",
            "    (0): Conv2d(16, 32, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
            "    (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    (2): ReLU()\n",
            "    (3): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
            "  )\n",
            "  (fc): Linear(in_features=1568, out_features=10, bias=True)\n",
            ")\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "ZpZoRX_AMSp3",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Construct loss function and optimizer\n",
        "criterion = nn.CrossEntropyLoss()                           # Cross Entropy Loss\n",
        "optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)  # Stochastic Gradient Descent"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "8I0PVCEvSYZg",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 718
        },
        "outputId": "a397daf4-a301-4458-c747-c7f5083ca5d4"
      },
      "cell_type": "code",
      "source": [
        "# Train the model\n",
        "losses = []\n",
        "\n",
        "# Train the model\n",
        "for epoch in range(num_epochs):\n",
        "    for i, (images, labels) in enumerate(train_loader):\n",
        "        # Move tensors to the configured device\n",
        "        images = images.to(device)\n",
        "        labels = labels.to(device)\n",
        "        \n",
        "        # Forward pass\n",
        "        outputs = model(images)\n",
        "        \n",
        "        # Compute loss\n",
        "        loss = criterion(outputs, labels)\n",
        "    \n",
        "        # Compute gradient and update model parameters\n",
        "        optimizer.zero_grad()\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "        \n",
        "    losses.append(loss.item())\n",
        "    print(\"Epoch:\", epoch, \"loss:\", loss.item())\n",
        "\n",
        "plt.plot(range(num_epochs), losses)\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('epoch')"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 0 loss: 0.05155560001730919\n",
            "Epoch: 1 loss: 0.054467249661684036\n",
            "Epoch: 2 loss: 0.012793012894690037\n",
            "Epoch: 3 loss: 0.08530494570732117\n",
            "Epoch: 4 loss: 0.04401886835694313\n",
            "Epoch: 5 loss: 0.008303243666887283\n",
            "Epoch: 6 loss: 0.024211039766669273\n",
            "Epoch: 7 loss: 0.0028443573974072933\n",
            "Epoch: 8 loss: 0.002394263166934252\n",
            "Epoch: 9 loss: 0.008879953995347023\n",
            "Epoch: 10 loss: 0.003187580034136772\n",
            "Epoch: 11 loss: 0.0011781537905335426\n",
            "Epoch: 12 loss: 0.0050160945393145084\n",
            "Epoch: 13 loss: 0.0011074519716203213\n",
            "Epoch: 14 loss: 0.016137603670358658\n",
            "Epoch: 15 loss: 0.00044471025466918945\n",
            "Epoch: 16 loss: 0.010700857266783714\n",
            "Epoch: 17 loss: 0.0016525005921721458\n",
            "Epoch: 18 loss: 0.010068688541650772\n",
            "Epoch: 19 loss: 0.0001672506332397461\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5,0,'epoch')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFYCAYAAABKymUhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fb4bec3c0b8>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "s8R4fwyKSaWn",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "bf96704a-1c9a-4edf-d38f-ea826f919548"
      },
      "cell_type": "code",
      "source": [
        "# Test the model\n",
        "model.eval()\n",
        "with torch.no_grad():\n",
        "    correct = 0\n",
        "    total = 0\n",
        "    for images, labels in train_loader:\n",
        "        images = images.to(device)\n",
        "        labels = labels.to(device)\n",
        "        outputs = model(images)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total += labels.size(0)\n",
        "        correct += (predicted == labels).sum().item()\n",
        "\n",
        "    print('Test Accuracy of the model on the ', len(train_dataset), ' train images: {} %'.format(100 * correct / total))\n",
        "    \n",
        "    correct = 0\n",
        "    total = 0\n",
        "    for images, labels in test_loader:\n",
        "        images = images.to(device)\n",
        "        labels = labels.to(device)\n",
        "        outputs = model(images)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total += labels.size(0)\n",
        "        correct += (predicted == labels).sum().item()\n",
        "\n",
        "    print('Test Accuracy of the model on the ', len(test_dataset), ' test images: {} %'.format(100 * correct / total))\n"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Test Accuracy of the model on the  60000  train images: 99.81 %\n",
            "Test Accuracy of the model on the  10000  test images: 99.09 %\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "HGz8GDfQo6zo",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}