Challenge_1_Final_Code.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HaFUp-OZ0a4T"
      },
      "source": [
        "Code Devised From -\n",
        "https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aBRhbx4y3o4k"
      },
      "source": [
        "## Import libraries"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y_3sxsX3Vwzy"
      },
      "source": [
        "For Adilina - dataset is in UdG Account\n",
        "\n",
        "For Anwai - dataset is in UdG Account"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "A9KLFTlhE3es",
        "outputId": "c9b10263-52c4-48c2-ad75-fd1715f1dbf9"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Collecting barbar\n",
            "  Downloading barbar-0.2.1-py3-none-any.whl (3.9 kB)\n",
            "Installing collected packages: barbar\n",
            "Successfully installed barbar-0.2.1\n",
            "Collecting efficientnet-pytorch\n",
            "  Downloading efficientnet_pytorch-0.7.1.tar.gz (21 kB)\n",
            "Requirement already satisfied: torch in /usr/local/lib/python3.7/dist-packages (from efficientnet-pytorch) (1.10.0+cu111)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torch->efficientnet-pytorch) (3.10.0.2)\n",
            "Building wheels for collected packages: efficientnet-pytorch\n",
            "  Building wheel for efficientnet-pytorch (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for efficientnet-pytorch: filename=efficientnet_pytorch-0.7.1-py3-none-any.whl size=16446 sha256=e061d9f2f6fc4a207bffd00b53903ea4b828568a36e0e9875290189017de590e\n",
            "  Stored in directory: /root/.cache/pip/wheels/0e/cc/b2/49e74588263573ff778da58cc99b9c6349b496636a7e165be6\n",
            "Successfully built efficientnet-pytorch\n",
            "Installing collected packages: efficientnet-pytorch\n",
            "Successfully installed efficientnet-pytorch-0.7.1\n",
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "import os\n",
        "import glob\n",
        "import random\n",
        "import skimage\n",
        "import imageio\n",
        "import matplotlib\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from PIL import Image\n",
        "import seaborn as sns\n",
        "from datetime import datetime\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "import scipy.ndimage\n",
        "from scipy import misc, stats\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import LabelEncoder, StandardScaler\n",
        "from sklearn.metrics import confusion_matrix, precision_recall_fscore_support, accuracy_score\n",
        "\n",
        "import torch\n",
        "import torchvision\n",
        "from torch import nn\n",
        "import torch.optim as optim\n",
        "import torch.nn.functional as F\n",
        "import torchvision.transforms as transforms\n",
        "\n",
        "%matplotlib inline\n",
        "%pip install barbar \n",
        "%pip install --upgrade efficientnet-pytorch\n",
        "from barbar import Bar\n",
        "\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "device\n",
        "\n",
        "from google.colab import drive\n",
        "drive._mount('/content/drive')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uquU7jReFmSc"
      },
      "source": [
        "## Configuration Options and Hyper Parameters"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ioy1TTbJFevo"
      },
      "outputs": [],
      "source": [
        "# Directories of the HAM10000 Dataset \n",
        "main_dir  = '/content/drive/MyDrive/CAD/Project - 2/Challenge 2/'\n",
        "train_dir = main_dir + 'data/train/'\n",
        "val_dir   = main_dir + 'data/val/'\n",
        "# test_dir  = main_dir + 'data/test/'\n",
        "\n",
        "# Classes of the Challenge 1\n",
        "classes   = ['nv', 'ls']\n",
        "\n",
        "# Normalization Mean and Standard Deviation Values for the Pretrained Weights on Imagenet Dataset\n",
        "norm_mean = (0.485, 0.456, 0.406)\n",
        "norm_std  = (0.229, 0.224, 0.225)\n",
        "\n",
        "# Hyperparameters\n",
        "num_size              = n # where n is the batch size you want, either the same for all the datasets or your individual choices.\n",
        "batch_size            = num_size\n",
        "validation_batch_size = num_size\n",
        "test_batch_size       = num_size\n",
        "\n",
        "num_epochs            = 100\n",
        "num_classes           = len(classes)\n",
        "learning_rate         = 1e-5\n",
        "patience              = 10\n",
        "\n",
        "# Image Size\n",
        "height                = 450\n",
        "width                 = 600\n",
        "\n",
        "# Dimension of the CenterCrop Transformation (448,224,256)\n",
        "centercrop            = c # depends on the configuration you prefer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K_eUcMLVF3co"
      },
      "source": [
        "## Data Augmentation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "K5L8-TDoF0uV"
      },
      "outputs": [],
      "source": [
        "# Specifying the Transformations for the Training Images\n",
        "transform_train = transforms.Compose([\n",
        "                    transforms.Resize((height,width)),\n",
        "                    transforms.CenterCrop(centercrop),\n",
        "                    transforms.RandomAffine(90,shear=20,scale=(0.8,1.2)),\n",
        "                    transforms.RandomVerticalFlip(),\n",
        "                    transforms.RandomHorizontalFlip(),\n",
        "                    transforms.ColorJitter(brightness=(0.7,1.3), contrast=(0.7,1.3), saturation=(0.7,1.3)),\n",
        "                    # transforms.RandomRotation(degrees=90),\n",
        "                    # transforms.RandomInvert(),\n",
        "                    transforms.ToTensor(),\n",
        "                    transforms.Normalize(norm_mean, norm_std),\n",
        "                    ])\n",
        "\n",
        "# Specifying the Transformations for the Validation Images\n",
        "transform_val = transforms.Compose([\n",
        "                    transforms.Resize((height,width)),\n",
        "                    transforms.CenterCrop(centercrop),\n",
        "                    transforms.ToTensor(),\n",
        "                    transforms.Normalize(norm_mean, norm_std),\n",
        "                    ])\n",
        "\n",
        "# Specifying the Transformations for the Testing Images\n",
        "transform_test = transforms.Compose([\n",
        "                    transforms.Resize((height,width)),\n",
        "                    transforms.CenterCrop(centercrop),\n",
        "                    transforms.ToTensor(),\n",
        "                    transforms.Normalize(norm_mean, norm_std),\n",
        "                    ])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oeoM6dSnF8Er"
      },
      "source": [
        "Data Generators"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "UD4WKvPIF6Md"
      },
      "outputs": [],
      "source": [
        "# Datasets\n",
        "traindataset = torchvision.datasets.ImageFolder(root = train_dir, transform = transform_train)\n",
        "valdataset   = torchvision.datasets.ImageFolder(root = val_dir, transform = transform_val)\n",
        "# testdataset  = torchvision.datasets.ImageFolder(root = test_dir, transform = transform_test)\n",
        "\n",
        "# Dataloaders\n",
        "train_data_loader      = torch.utils.data.DataLoader(traindataset, batch_size = batch_size, shuffle = True)\n",
        "validation_data_loader = torch.utils.data.DataLoader(valdataset, batch_size = validation_batch_size, shuffle = False)\n",
        "# test_data_loader       = torch.utils.data.DataLoader(testdataset, batch_size = test_batch_size, shuffle = False)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0bzCHyM9GAbG"
      },
      "source": [
        "Visualising the Transformation Images"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "id": "0OwDo-3dF-ZW",
        "outputId": "250604f0-3a7a-442f-ede9-cedde4e41371"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "   nv,     nv,     nv, \n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x1080 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Function to Show Image\n",
        "fig = plt.figure(figsize=(10, 15))\n",
        "def imshow(img):\n",
        "    img = img / 2 + 0.5     # unnormalise\n",
        "    npimg = img.numpy()\n",
        "    plt.axis('off')\n",
        "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
        "\n",
        "# Obtain Random Training Images\n",
        "dataiter       = iter(validation_data_loader)\n",
        "images, labels = dataiter.next()\n",
        "\n",
        "# Show the Images\n",
        "imshow(torchvision.utils.make_grid(images))\n",
        "\n",
        "print(' '.join('%5s, ' % classes[labels[j]] for j in range(len(labels))))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nbmm5fdkGHyk"
      },
      "source": [
        "## Architecture"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UP2mXOtdWaHx",
        "outputId": "8b386673-14bb-438a-e23e-81c2bb15b3ea"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "GoogLeNet(\n",
            "  (conv1): BasicConv2d(\n",
            "    (conv): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
            "    (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "  )\n",
            "  (maxpool1): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (conv2): BasicConv2d(\n",
            "    (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "    (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "  )\n",
            "  (conv3): BasicConv2d(\n",
            "    (conv): Conv2d(64, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "    (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "  )\n",
            "  (maxpool2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (inception3a): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(96, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(192, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(16, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(192, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception3b): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(128, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(256, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (maxpool3): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (inception4a): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(480, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(480, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(96, 208, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(208, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(480, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(16, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(16, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(480, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4b): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(512, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(112, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(112, 224, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(224, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(24, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4c): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(24, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4d): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(512, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(112, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(144, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(144, 288, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(288, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4e): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(528, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(528, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(160, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(320, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(528, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(528, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (maxpool4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (inception5a): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(832, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(160, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(320, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(832, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception5b): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(832, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(384, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(384, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(48, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(832, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (aux1): None\n",
            "  (aux2): None\n",
            "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
            "  (dropout): Dropout(p=0.2, inplace=False)\n",
            "  (fc): Linear(in_features=1024, out_features=1000, bias=True)\n",
            ")\n"
          ]
        }
      ],
      "source": [
        "# Viewing the Architecture and Finding the Cross-Section of Transfer Learning\n",
        "dummy_net = torchvision.models.googlenet(pretrained = True) \n",
        "print(dummy_net)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7J_XL-cCWcKp",
        "outputId": "a098287a-0b1a-4dcb-e4dd-69d4b68364c5"
      },
      "outputs": [
        {
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'(classifier): Sequential(\\n    (0): Dropout(p=0.5, inplace=False)\\n    (1): Linear(in_features=9216, out_features=4096, bias=True)\\n    (2): ReLU(inplace=True)\\n    (3): Dropout(p=0.5, inplace=False)\\n    (4): Linear(in_features=4096, out_features=4096, bias=True)\\n    (5): ReLU(inplace=True)\\n    (6): Linear(in_features=4096, out_features=1000, bias=True)'"
            ]
          },
          "execution_count": 18,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Alexnet Final Layers Sample - How to Play with them in the Necessary Structure\n",
        "\"\"\"(classifier): Sequential(\n",
        "    (0): Dropout(p=0.5, inplace=False)\n",
        "    (1): Linear(in_features=9216, out_features=4096, bias=True)\n",
        "    (2): ReLU(inplace=True)\n",
        "    (3): Dropout(p=0.5, inplace=False)\n",
        "    (4): Linear(in_features=4096, out_features=4096, bias=True)\n",
        "    (5): ReLU(inplace=True)\n",
        "    (6): Linear(in_features=4096, out_features=1000, bias=True)\"\"\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7Y5z5KI8aPoz",
        "outputId": "01a0d97b-62da-4578-f3fd-7221d2744e57"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "GoogLeNet(\n",
            "  (conv1): BasicConv2d(\n",
            "    (conv): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
            "    (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "  )\n",
            "  (maxpool1): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (conv2): BasicConv2d(\n",
            "    (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "    (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "  )\n",
            "  (conv3): BasicConv2d(\n",
            "    (conv): Conv2d(64, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "    (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "  )\n",
            "  (maxpool2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (inception3a): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(96, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(192, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(16, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(192, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception3b): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(128, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(256, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (maxpool3): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (inception4a): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(480, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(480, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(96, 208, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(208, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(480, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(16, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(16, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(480, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4b): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(512, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(112, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(112, 224, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(224, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(24, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4c): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(24, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4d): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(512, 112, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(112, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(144, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(144, 288, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(288, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception4e): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(528, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(528, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(160, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(320, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(528, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(528, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (maxpool4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=True)\n",
            "  (inception5a): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(832, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(160, 320, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(320, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(32, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(32, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(832, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (inception5b): Inception(\n",
            "    (branch1): BasicConv2d(\n",
            "      (conv): Conv2d(832, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "      (bn): BatchNorm2d(384, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "    )\n",
            "    (branch2): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(384, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch3): Sequential(\n",
            "      (0): BasicConv2d(\n",
            "        (conv): Conv2d(832, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(48, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "    (branch4): Sequential(\n",
            "      (0): MaxPool2d(kernel_size=3, stride=1, padding=1, dilation=1, ceil_mode=True)\n",
            "      (1): BasicConv2d(\n",
            "        (conv): Conv2d(832, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
            "        (bn): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (aux1): None\n",
            "  (aux2): None\n",
            "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
            "  (dropout): Dropout(p=0.2, inplace=False)\n",
            "  (fc): Linear(in_features=1024, out_features=2, bias=True)\n",
            ")\n",
            "/content/drive/MyDrive/CAD/Project - 2/Challenge 1/pytorch-binary-models/try-cad1-checkpoint-epoch100-batchsize2-googlenet.pt\n"
          ]
        }
      ],
      "source": [
        "# The Choice of our Architecture to Train\n",
        "model = \"googlenet\"\n",
        "\n",
        "# Replacing the Last Layer(s) of \"x\"net to Match our Number of Classes\n",
        "net    = torchvision.models.googlenet(pretrained = True)\n",
        "net.fc = nn.Linear(1024,num_classes)\n",
        "\n",
        "print(net)\n",
        "net = net.to(device)\n",
        "\n",
        "# Number of Loops over the Dataset\n",
        "model_save_dir = main_dir + 'pytorch-binary-models/try-cad1-checkpoint-epoch{0}-batchsize{1}-{2}.pt'.format(num_epochs, num_size, str(model))\n",
        "print(model_save_dir)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "\"\"\"# If the model is to be used as a Feature Extractor, setting its parameters to non-trainable (by default, trainable)\n",
        "for param in model.parameters():\n",
        "    param.requires_grad = False\"\"\""
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IwCGjZFNGPmp"
      },
      "source": [
        "## Functions for Metrics and Validation Evaluation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "zwQwPjBMGJxd"
      },
      "outputs": [],
      "source": [
        "def get_accuracy(predicted, labels):\n",
        "    batch_len, correct= 0, 0\n",
        "    batch_len = labels.size(0)\n",
        "    correct = (predicted == labels).sum().item()\n",
        "    return batch_len, correct\n",
        "\n",
        "def evaluate(model, val_loader):\n",
        "    losses= 0\n",
        "    num_samples_total=0\n",
        "    correct_total=0\n",
        "    model.eval()\n",
        "    for inputs, labels in val_loader:\n",
        "        inputs, labels = inputs.to(device), labels.to(device)\n",
        "        out = model(inputs)\n",
        "        _, predicted = torch.max(out, 1)\n",
        "        loss = criterion(out, labels)\n",
        "        losses += loss.item() \n",
        "        b_len, corr = get_accuracy(predicted, labels)\n",
        "        num_samples_total +=b_len\n",
        "        correct_total +=corr\n",
        "    accuracy = correct_total/num_samples_total\n",
        "    losses = losses/len(val_loader)\n",
        "    return losses, accuracy\n",
        "    \n",
        "class EarlyStopping:\n",
        "    \"\"\"Early Stops the Training if the Validation Loss doesn't improve after a predefined Patience.\"\"\"\n",
        "    def __init__(self, patience=7, verbose=False, delta=0):\n",
        "\n",
        "        self.patience     = patience\n",
        "        self.verbose      = verbose\n",
        "        self.counter      = 0\n",
        "        self.best_score   = None\n",
        "        self.early_stop   = False\n",
        "        self.val_loss_min = np.Inf\n",
        "        self.delta        = delta\n",
        "\n",
        "    def __call__(self, val_loss, model):\n",
        "\n",
        "        score = -val_loss\n",
        "\n",
        "        if self.best_score is None:\n",
        "            self.best_score = score\n",
        "            self.save_checkpoint(val_loss, model)\n",
        "        elif score < self.best_score + self.delta:\n",
        "            self.counter += 1\n",
        "            print(f'EarlyStopping Counter: {self.counter} out of {self.patience}')\n",
        "            if self.counter >= self.patience:\n",
        "                self.early_stop = True\n",
        "        else:\n",
        "            self.best_score = score\n",
        "            self.save_checkpoint(val_loss, model)\n",
        "            self.counter = 0\n",
        "\n",
        "    def save_checkpoint(self, val_loss, model):\n",
        "        '''Saves model when validation loss decrease.'''\n",
        "        if self.verbose:\n",
        "            print(f'Validation Loss Decreased ({self.val_loss_min:.6f} --> {val_loss:.6f}).  Saving model ...')\n",
        "        torch.save(model.state_dict(), model_save_dir)\n",
        "        self.val_loss_min = val_loss"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qtR0iHGbGUCf"
      },
      "source": [
        "## Training Progress"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "VO-NeM6YH-ss"
      },
      "outputs": [],
      "source": [
        "# Clearing the GPU Cache\n",
        "import torch\n",
        "torch.cuda.empty_cache()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nA7Mk1wQyQGQ",
        "outputId": "88d50465-65fd-472d-f161-604537d2ae87"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Starting Training with Network: googlenet, Loading Checkpoint Status : False\n"
          ]
        }
      ],
      "source": [
        "load_checkpoint  = False\n",
        "print(\"Starting Training with Network: \" + model + \", Loading Checkpoint Status : \" + str(load_checkpoint))\n",
        "\n",
        "# Optimizer and Loss Configurations\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = optim.Adam(net.parameters(), lr=learning_rate)\n",
        "scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(optimizer,'min')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9Qdib0A1GSSO",
        "outputId": "b1075392-f382-457d-b343-751caef6f696"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "16000/16000: [===============================>] - ETA 0.9s\n",
            "Epoch: 1\n",
            "Training Loss: 0.564 , Training Accuracy:0.704 => Validation Loss: 0.496, Validation Accuracy: 0.755 \n",
            "Validation loss decreased (inf --> 0.495890).  Saving model ...\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 2\n",
            "Training Loss: 0.401 , Training Accuracy:0.817 => Validation Loss: 0.342, Validation Accuracy: 0.837 \n",
            "Validation loss decreased (0.495890 --> 0.341688).  Saving model ...\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 3\n",
            "Training Loss: 0.364 , Training Accuracy:0.835 => Validation Loss: 0.309, Validation Accuracy: 0.849 \n",
            "Validation loss decreased (0.341688 --> 0.309080).  Saving model ...\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 4\n",
            "Training Loss: 0.341 , Training Accuracy:0.844 => Validation Loss: 0.295, Validation Accuracy: 0.866 \n",
            "Validation loss decreased (0.309080 --> 0.294643).  Saving model ...\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 5\n",
            "Training Loss: 0.327 , Training Accuracy:0.850 => Validation Loss: 0.295, Validation Accuracy: 0.858 \n",
            "EarlyStopping counter: 1 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 6\n",
            "Training Loss: 0.312 , Training Accuracy:0.858 => Validation Loss: 0.273, Validation Accuracy: 0.876 \n",
            "Validation loss decreased (0.294643 --> 0.272686).  Saving model ...\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 7\n",
            "Training Loss: 0.300 , Training Accuracy:0.863 => Validation Loss: 0.281, Validation Accuracy: 0.870 \n",
            "EarlyStopping counter: 1 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 8\n",
            "Training Loss: 0.292 , Training Accuracy:0.871 => Validation Loss: 0.271, Validation Accuracy: 0.880 \n",
            "Validation loss decreased (0.272686 --> 0.270504).  Saving model ...\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 9\n",
            "Training Loss: 0.281 , Training Accuracy:0.874 => Validation Loss: 0.302, Validation Accuracy: 0.870 \n",
            "EarlyStopping counter: 1 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 10\n",
            "Training Loss: 0.271 , Training Accuracy:0.883 => Validation Loss: 0.279, Validation Accuracy: 0.886 \n",
            "EarlyStopping counter: 2 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 11\n",
            "Training Loss: 0.261 , Training Accuracy:0.885 => Validation Loss: 0.280, Validation Accuracy: 0.878 \n",
            "EarlyStopping counter: 3 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 12\n",
            "Training Loss: 0.250 , Training Accuracy:0.890 => Validation Loss: 0.276, Validation Accuracy: 0.884 \n",
            "EarlyStopping counter: 4 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 13\n",
            "Training Loss: 0.248 , Training Accuracy:0.892 => Validation Loss: 0.297, Validation Accuracy: 0.876 \n",
            "EarlyStopping counter: 5 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 14\n",
            "Training Loss: 0.236 , Training Accuracy:0.897 => Validation Loss: 0.300, Validation Accuracy: 0.870 \n",
            "EarlyStopping counter: 6 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 15\n",
            "Training Loss: 0.227 , Training Accuracy:0.901 => Validation Loss: 0.299, Validation Accuracy: 0.865 \n",
            "EarlyStopping counter: 7 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 16\n",
            "Training Loss: 0.217 , Training Accuracy:0.904 => Validation Loss: 0.282, Validation Accuracy: 0.876 \n",
            "EarlyStopping counter: 8 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 17\n",
            "Training Loss: 0.210 , Training Accuracy:0.911 => Validation Loss: 0.292, Validation Accuracy: 0.882 \n",
            "EarlyStopping counter: 9 out of 10\n",
            "16000/16000: [===============================>] - ETA 0.1s\n",
            "Epoch: 18\n",
            "Training Loss: 0.201 , Training Accuracy:0.912 => Validation Loss: 0.293, Validation Accuracy: 0.887 \n",
            "EarlyStopping counter: 10 out of 10\n",
            "Early Stopping\n",
            "Finished Training\n"
          ]
        }
      ],
      "source": [
        "if load_checkpoint == False:\n",
        "  #net.load_state_dict(torch.load('/content/drive/MyDrive/CAD/Project - 2/Challenge 2/pytorch-binary-models/checkpoint-epoch-{0}.pt'.format(num_epochs)))\n",
        "  #print(\"Here\")\n",
        "  \n",
        "  # Initializing the early_stopping\n",
        "  early_stopping = EarlyStopping(patience=patience, verbose=True)\n",
        "\n",
        "  accuracy      = []\n",
        "  val_accuracy  = []\n",
        "  losses        = []\n",
        "  val_losses    = []\n",
        "  best_accuracy = 0\n",
        "\n",
        "  for epoch in range(num_epochs):\n",
        "      running_loss      = 0.0\n",
        "      correct_total     = 0.0\n",
        "      num_samples_total = 0.0\n",
        "      for i, data in enumerate(Bar(train_data_loader)):\n",
        "          # Getting the Inputs\n",
        "          inputs, labels = data\n",
        "          inputs, labels = inputs.to(device), labels.to(device)\n",
        "\n",
        "          # Setting the Parameter Gradients to Zero\n",
        "          optimizer.zero_grad()\n",
        "\n",
        "          # Forward Pass + Backward Pass + Optimize\n",
        "          outputs = net(inputs)\n",
        "          loss    = criterion(outputs, labels)\n",
        "          loss.backward()\n",
        "          optimizer.step()\n",
        "          \n",
        "          # Computing Accuracy\n",
        "          _, predicted = torch.max(outputs, 1)\n",
        "          b_len, corr = get_accuracy(predicted, labels)\n",
        "          num_samples_total +=b_len\n",
        "          correct_total +=corr\n",
        "          running_loss += loss.item()\n",
        "\n",
        "      running_loss /= len(train_data_loader)\n",
        "      train_accuracy = correct_total/num_samples_total\n",
        "      val_loss, val_acc = evaluate(net, validation_data_loader)\n",
        "      scheduler.step(val_loss)\n",
        "      \n",
        "      print('Epoch: %d' %(epoch+1))\n",
        "      print('Training Loss: %.3f , Training Accuracy:%.3f => Validation Loss: %.3f, Validation Accuracy: %.3f ' %(running_loss, train_accuracy,val_loss, val_acc))\n",
        "\n",
        "      losses.append(running_loss)     # List Containing the Training Losses\n",
        "      val_losses.append(val_loss)     # List Containing the Validation Losses\n",
        "      accuracy.append(train_accuracy) # List Containing the Training Accuracy\n",
        "      val_accuracy.append(val_acc)    # List Containing the Validation Accuracy\n",
        "  \n",
        "      # early_stopping needs Validation Loss to check if it decreases and if so, it makes a Checkpoint of the Current Model\n",
        "      early_stopping(val_loss, net)\n",
        "      \n",
        "      if early_stopping.early_stop:\n",
        "          print(\"Early Stopping\")\n",
        "          break\n",
        "\n",
        "  print('Finished Training')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pyix8nUeI3hN"
      },
      "source": [
        "Training and Validation Accuracy Curve"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "20YNPFtBHodO",
        "outputId": "352c972a-6c93-4ebf-d038-8d05061c03fa"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fc9803dc490>"
            ]
          },
          "execution_count": 32,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "#tmpdir = \"/content/\"\n",
        "plt.figure()\n",
        "plt.plot(accuracy)\n",
        "plt.plot(val_accuracy)\n",
        "plt.title('Accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epochs')\n",
        "plt.legend(['train', 'validation'], loc='upper left')\n",
        "#plt.savefig(\"{}/accuracy.png\".format(tmpdir))\n",
        "\n",
        "# Training and Validation Loss\n",
        "plt.figure()\n",
        "plt.plot(losses)\n",
        "plt.plot(val_losses)\n",
        "plt.title('Loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epochs')\n",
        "plt.legend(['train', 'validation'], loc='upper left')\n",
        "#plt.savefig(\"{}/loss.png\".format(tmpdir))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eEJjBYvJJC1l"
      },
      "source": [
        "## Testing Accuracy and Confusion Matrix\n",
        "Validation Accuracy and Confusion Matrix without Test Time Augmentation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xf8MAgaeWgEi",
        "outputId": "d4046314-9693-4627-f19a-8fbd3aaaa59b"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<All keys matched successfully>"
            ]
          },
          "execution_count": 25,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Loading the Best Model\n",
        "net.load_state_dict(torch.load(model_save_dir))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9t746QnGI_0T",
        "outputId": "56198cb4-9434-4e71-b288-f7b45e4093db"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Accuracy of the Network on the Validation Images: 88 %\n"
          ]
        }
      ],
      "source": [
        "#net.load_state_dict(torch.load('/content/drive/MyDrive/CAD/Project - 2/Challenge 2/pytorch-multiclass-models/checkpoint-epoch-{0}.pt'.format(num_epochs)))\n",
        "\n",
        "correct = 0\n",
        "total   = 0\n",
        "net.eval()\n",
        "\n",
        "with torch.no_grad():\n",
        "    for data in validation_data_loader:\n",
        "        images, labels = data\n",
        "        images, labels = images.to(device), labels.to(device)\n",
        "        outputs = net(images)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total += labels.size(0)\n",
        "        correct += (predicted == labels).sum().item()\n",
        "\n",
        "print('Accuracy of the Network on the Validation Images: %d %%' % (\n",
        "    100 * correct / total))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-tvMgAWg8AJ3",
        "outputId": "bcbe465c-f292-4364-a3d9-517a28bd277a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "tensor([[538.,  62.],\n",
            "        [ 82., 518.]]) \n",
            "\n",
            "1200.0\n",
            "nv - recall :  0.86774194  precision :  0.89666665  accuracy :  0.88\n",
            "ls - recall :  0.8931034  precision :  0.86333334  accuracy :  0.88\n"
          ]
        }
      ],
      "source": [
        "confusion_matrix = torch.zeros(len(classes), len(classes))\n",
        "with torch.no_grad():\n",
        "    for data in validation_data_loader:\n",
        "        images, labels = data\n",
        "        images, labels = images.to(device), labels.to(device)\n",
        "        outputs = net(images)\n",
        "        _, predicted = torch.max(outputs, 1)\n",
        "        for t, p in zip(labels.view(-1), predicted.view(-1)):\n",
        "                confusion_matrix[t.long(), p.long()] += 1\n",
        "\n",
        "print(confusion_matrix, \"\\n\")\n",
        "cm = confusion_matrix.numpy()\n",
        "\n",
        "recall_nv = cm[0][0]/np.sum(cm[:,0])\n",
        "recall_ls = cm[1][1]/np.sum(cm[:,1])\n",
        "\n",
        "precision_nv = cm[0][0]/np.sum(cm[0])\n",
        "precision_ls = cm[1][1]/np.sum(cm[1])\n",
        "\n",
        "total = np.sum(cm)\n",
        "print(total)\n",
        "accuracy_nv = cm[0][0] + (total - np.sum(cm[0]) - np.sum(cm[:,0]) + cm[0][0])\n",
        "accuracy_ls = cm[1][1] + (total - np.sum(cm[1]) - np.sum(cm[:,1]) + cm[1][1])\n",
        "\n",
        "print(\"nv - recall : \",recall_nv, \" precision : \",precision_nv, \" accuracy : \",(accuracy_nv/total))\n",
        "print(\"ls - recall : \",recall_ls, \" precision : \",precision_ls, \" accuracy : \",(accuracy_ls/total))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 503
        },
        "id": "38gUjkwZ8Dk9",
        "outputId": "342f9721-976d-4526-fff1-643e77701e5a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "0.88\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 504x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig,ax= plt.subplots(figsize=(7,7))\n",
        "sns.heatmap(cm / (cm.astype(np.float).sum(axis=1) + 1e-9), annot=False, ax=ax)\n",
        "\n",
        "# Labels, Title and Ticks\n",
        "ax.set_xlabel('Predicted', size=25);\n",
        "ax.set_ylabel('True', size=25); \n",
        "ax.set_title('Confusion Matrix', size=25); \n",
        "ax.xaxis.set_ticklabels(['nv','ls'], size=15); \n",
        "ax.yaxis.set_ticklabels(['nv','ls'], size=15);\n",
        "print(correct/total)"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "machine_shape": "hm",
      "name": "Challenge 1 - Final Code.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}