Ivy__Deep_Cars_Identifying_Car_Brands.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python",
      "version": "3.6.4",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "colab": {
      "name": "Ivy_ Deep Cars - Identifying Car Brands.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ivyclare/DeepCars---Transfer-Learning-With-Pytorch/blob/master/Ivy__Deep_Cars_Identifying_Car_Brands.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
        "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
        "trusted": true,
        "id": "tW_TpN633LTM",
        "colab_type": "code",
        "outputId": "35f27c28-d2eb-42e9-ec2a-68908e9ed582",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 33
        }
      },
      "source": [
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.image as mpimg\n",
        "import seaborn as sns\n",
        "\n",
        "import torch\n",
        "from torchvision import datasets,transforms,models\n",
        "from torch import nn,optim\n",
        "import torch.nn.functional as F\n",
        "from torch.utils.data import *\n",
        "\n",
        "import time\n",
        "import json\n",
        "import copy\n",
        "import os\n",
        "import glob\n",
        "\n",
        "from PIL import Image\n",
        "\n",
        "#os.environ['CUDA_LAUNCH_BLOCKING'] = \"1\"\n",
        "torch.backends.cudnn.enabled = False\n",
        "# check if CUDA is available\n",
        "train_on_gpu = torch.cuda.is_available()\n",
        "# torch.backends.cudnn.enabled\n",
        "\n",
        "if not train_on_gpu:\n",
        "    print('CUDA is not available.  Training on CPU ...')\n",
        "else:\n",
        "    print('CUDA is available!  Training on GPU ...')"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CUDA is available!  Training on GPU ...\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ibvAG1Gc3UDo",
        "colab_type": "code",
        "outputId": "7bdfa573-8522-4eb1-901c-8b4439e8ba2d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 33
        }
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SIZa1EfL3fu4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Downloading the data from Kaggle\n",
        "\n",
        "!pip install kaggle\n",
        "!pip install -q kaggle\n",
        "\n",
        "!mkdir .kaggle\n",
        "\n",
        "import json\n",
        "token = {\"username\":\"ivyclare\",\"key\":\"17ee8bd3b41486d62e7eb9257bd812d4\"}\n",
        "with open('/content/.kaggle/kaggle.json', 'w') as file:\n",
        "  json.dump(token, file)\n",
        "\n",
        "!chmod 600 /content/.kaggle/kaggle.json\n",
        "\n",
        "!cp /content/.kaggle/kaggle.json ~/.kaggle/kaggle.json\n",
        " \n",
        "!kaggle config set -n path -v{/content}\n",
        "\n",
        "#!kaggle datasets download -d spaics/hackathon-blossom-flower-classification -p /content\n",
        "#!kaggle competitions download -c dog-breed-identification -p /content\n",
        "!kaggle competitions download -c virtual-hack -p /content\n",
        "#!kaggle competitions download -c dog-breed-identification -p /content\n",
        "!unzip \\*.zip\n",
        "\n",
        "# #!unzip \\*.zip"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
        "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a",
        "trusted": true,
        "id": "uWonxtay3LTU",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Tansform with data augmentation and normalization for training\n",
        "# Just normalization for validation\n",
        "data_transforms = {\n",
        "    'train': transforms.Compose([\n",
        "        transforms.RandomRotation(30),\n",
        "        transforms.RandomResizedCrop(224),\n",
        "        #transforms.RandomResizedCrop(299),\n",
        "        transforms.RandomHorizontalFlip(),\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
        "    ]),\n",
        "    'valid': transforms.Compose([\n",
        "        #transforms.Resize(256),\n",
        "        transforms.CenterCrop(224),\n",
        "        transforms.CenterCrop(299),\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
        "    ]),\n",
        "    \n",
        "    'test': transforms.Compose([\n",
        "        transforms.Resize(256),                                  \n",
        "        transforms.CenterCrop(224),\n",
        "        #transforms.CenterCrop(299),\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Normalize([0.485, 0.456, 0.406],[0.229, 0.224, 0.225])\n",
        "    ]),\n",
        "}\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "rGKCXVNb3LTX",
        "colab_type": "code",
        "outputId": "0820eda8-fde0-4002-e6aa-3a65f4591eeb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "label_df = pd.read_csv('names.csv', names=[\"label\"])\n",
        "label_df.head()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>AM General Hummer SUV 2000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Acura RL Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Acura TL Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Acura TL Type-S 2008</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Acura TSX Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>Acura Integra Type R 2001</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Acura ZDX Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Aston Martin V8 Vantage Convertible 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Aston Martin V8 Vantage Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Aston Martin Virage Convertible 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>Aston Martin Virage Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>Audi RS 4 Convertible 2008</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>Audi A5 Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>Audi TTS Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>Audi R8 Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>Audi V8 Sedan 1994</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>Audi 100 Sedan 1994</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>Audi 100 Wagon 1994</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>Audi TT Hatchback 2011</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>Audi S6 Sedan 2011</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>Audi S5 Convertible 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>Audi S5 Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>Audi S4 Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>Audi S4 Sedan 2007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>Audi TT RS Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>BMW ActiveHybrid 5 Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>BMW 1 Series Convertible 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>BMW 1 Series Coupe 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>BMW 3 Series Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>BMW 3 Series Wagon 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>166</th>\n",
              "      <td>Mitsubishi Lancer Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>167</th>\n",
              "      <td>Nissan Leaf Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>168</th>\n",
              "      <td>Nissan NV Passenger Van 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>169</th>\n",
              "      <td>Nissan Juke Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>170</th>\n",
              "      <td>Nissan 240SX Coupe 1998</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>171</th>\n",
              "      <td>Plymouth Neon Coupe 1999</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>172</th>\n",
              "      <td>Porsche Panamera Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>173</th>\n",
              "      <td>Ram C/V Cargo Van Minivan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>174</th>\n",
              "      <td>Rolls-Royce Phantom Drophead Coupe Convertible...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>175</th>\n",
              "      <td>Rolls-Royce Ghost Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>176</th>\n",
              "      <td>Rolls-Royce Phantom Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>177</th>\n",
              "      <td>Scion xD Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>178</th>\n",
              "      <td>Spyker C8 Convertible 2009</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>179</th>\n",
              "      <td>Spyker C8 Coupe 2009</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>180</th>\n",
              "      <td>Suzuki Aerio Sedan 2007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>181</th>\n",
              "      <td>Suzuki Kizashi Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>182</th>\n",
              "      <td>Suzuki SX4 Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>183</th>\n",
              "      <td>Suzuki SX4 Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>184</th>\n",
              "      <td>Tesla Model S Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>185</th>\n",
              "      <td>Toyota Sequoia SUV 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>186</th>\n",
              "      <td>Toyota Camry Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>187</th>\n",
              "      <td>Toyota Corolla Sedan 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>188</th>\n",
              "      <td>Toyota 4Runner SUV 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>189</th>\n",
              "      <td>Volkswagen Golf Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>190</th>\n",
              "      <td>Volkswagen Golf Hatchback 1991</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>191</th>\n",
              "      <td>Volkswagen Beetle Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>192</th>\n",
              "      <td>Volvo C30 Hatchback 2012</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>193</th>\n",
              "      <td>Volvo 240 Sedan 1993</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>194</th>\n",
              "      <td>Volvo XC90 SUV 2007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>195</th>\n",
              "      <td>smart fortwo Convertible 2012</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>196 rows × 1 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                 label\n",
              "0                           AM General Hummer SUV 2000\n",
              "1                                  Acura RL Sedan 2012\n",
              "2                                  Acura TL Sedan 2012\n",
              "3                                 Acura TL Type-S 2008\n",
              "4                                 Acura TSX Sedan 2012\n",
              "5                            Acura Integra Type R 2001\n",
              "6                             Acura ZDX Hatchback 2012\n",
              "7             Aston Martin V8 Vantage Convertible 2012\n",
              "8                   Aston Martin V8 Vantage Coupe 2012\n",
              "9                 Aston Martin Virage Convertible 2012\n",
              "10                      Aston Martin Virage Coupe 2012\n",
              "11                          Audi RS 4 Convertible 2008\n",
              "12                                  Audi A5 Coupe 2012\n",
              "13                                 Audi TTS Coupe 2012\n",
              "14                                  Audi R8 Coupe 2012\n",
              "15                                  Audi V8 Sedan 1994\n",
              "16                                 Audi 100 Sedan 1994\n",
              "17                                 Audi 100 Wagon 1994\n",
              "18                              Audi TT Hatchback 2011\n",
              "19                                  Audi S6 Sedan 2011\n",
              "20                            Audi S5 Convertible 2012\n",
              "21                                  Audi S5 Coupe 2012\n",
              "22                                  Audi S4 Sedan 2012\n",
              "23                                  Audi S4 Sedan 2007\n",
              "24                               Audi TT RS Coupe 2012\n",
              "25                       BMW ActiveHybrid 5 Sedan 2012\n",
              "26                       BMW 1 Series Convertible 2012\n",
              "27                             BMW 1 Series Coupe 2012\n",
              "28                             BMW 3 Series Sedan 2012\n",
              "29                             BMW 3 Series Wagon 2012\n",
              "..                                                 ...\n",
              "166                       Mitsubishi Lancer Sedan 2012\n",
              "167                         Nissan Leaf Hatchback 2012\n",
              "168                       Nissan NV Passenger Van 2012\n",
              "169                         Nissan Juke Hatchback 2012\n",
              "170                            Nissan 240SX Coupe 1998\n",
              "171                           Plymouth Neon Coupe 1999\n",
              "172                        Porsche Panamera Sedan 2012\n",
              "173                     Ram C/V Cargo Van Minivan 2012\n",
              "174  Rolls-Royce Phantom Drophead Coupe Convertible...\n",
              "175                       Rolls-Royce Ghost Sedan 2012\n",
              "176                     Rolls-Royce Phantom Sedan 2012\n",
              "177                            Scion xD Hatchback 2012\n",
              "178                         Spyker C8 Convertible 2009\n",
              "179                               Spyker C8 Coupe 2009\n",
              "180                            Suzuki Aerio Sedan 2007\n",
              "181                          Suzuki Kizashi Sedan 2012\n",
              "182                          Suzuki SX4 Hatchback 2012\n",
              "183                              Suzuki SX4 Sedan 2012\n",
              "184                           Tesla Model S Sedan 2012\n",
              "185                            Toyota Sequoia SUV 2012\n",
              "186                            Toyota Camry Sedan 2012\n",
              "187                          Toyota Corolla Sedan 2012\n",
              "188                            Toyota 4Runner SUV 2012\n",
              "189                     Volkswagen Golf Hatchback 2012\n",
              "190                     Volkswagen Golf Hatchback 1991\n",
              "191                   Volkswagen Beetle Hatchback 2012\n",
              "192                           Volvo C30 Hatchback 2012\n",
              "193                               Volvo 240 Sedan 1993\n",
              "194                                Volvo XC90 SUV 2007\n",
              "195                      smart fortwo Convertible 2012\n",
              "\n",
              "[196 rows x 1 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "Q-ybiuRr3LTa",
        "colab_type": "code",
        "outputId": "c0dceeeb-aed0-4feb-b34f-4be13bcc2d97",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 117
        }
      },
      "source": [
        "#Loading in the dataset\n",
        "\n",
        "train_dir = 'car_data/train'\n",
        "test_dir = 'car_data/test'\n",
        "label_dir = 'names.csv'\n",
        "\n",
        "batch_size=32\n",
        "dataset = datasets.ImageFolder(train_dir,transform=data_transforms['train'])\n",
        "\n",
        "# splitting our data\n",
        "valid_size  = int(0.1 * len(dataset))\n",
        "train_size = len(dataset) - valid_size\n",
        "dataset_sizes = {'train': train_size, 'valid': valid_size}\n",
        "\n",
        "# now we get our datasets\n",
        "train_dataset, valid_dataset = torch.utils.data.random_split(dataset, [train_size, valid_size])\n",
        "\n",
        "# Loading datasets into dataloader \n",
        "dataloaders = {'train': DataLoader(train_dataset, batch_size = batch_size, shuffle = True),\n",
        "              'valid': DataLoader(valid_dataset, batch_size = batch_size, shuffle = False)}\n",
        "\n",
        "print(\"Total Number of Samples: \",len(dataset))\n",
        "print(\"Number of Samples in Train: \",len(train_dataset))\n",
        "print(\"Number of Samples in Valid: \",len(valid_dataset))\n",
        "print(\"Number of Classes: \",len(dataset.classes))\n",
        "\n",
        "print(dataset.classes[0])"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "8144\n",
            "Total Number of Samples:  8144\n",
            "Number of Samples in Train:  7330\n",
            "Number of Samples in Valid:  814\n",
            "Number of Classes:  196\n",
            "AM General Hummer SUV 2000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_qnN1-Ws3LTe",
        "colab_type": "text"
      },
      "source": [
        "### Data Exploration and Visualization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "3d73bKS-3LTf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# ## Finding the number of samples per class (From https://www.kaggle.com/venkatarathnam/dog-breed-identification)\n",
        "# label_count_lis = [0]*len(dataset.classes)\n",
        "# label_count_lis = [0]*len(dataset.classes)\n",
        "# for classname, row in dataset.classes:\n",
        "#     label = dataset.classes.index.index(row['label'])\n",
        "#     label_count_lis[label] += 1\n",
        "# print (\"Minimum samples per class : {0}\".format(min(label_count_lis)))\n",
        "# print (\"Maximum samples per class : {0}\".format(max(label_count_lis)))\n",
        "\n",
        "# plt.figure()\n",
        "# f, ax = plt.subplots(figsize=(15,5))\n",
        "# plt.bar(np.arange(len(dataset.classes)), label_count_lis)\n",
        "# plt.xticks(np.arange(len(dataset.classes)), np.arange(len(dataset.classes)))\n",
        "# plt.ylabel(\"# of samples\")\n",
        "# plt.xlabel(\"Classes\")\n",
        "# plt.title('Train Set')\n",
        "# plt.show()\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "egy5xPoh3LTi",
        "colab_type": "code",
        "outputId": "608d42ac-8657-4189-b39c-71191a537137",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 239
        }
      },
      "source": [
        "## Method to display Image for Tensor\n",
        "def imshow(image, ax=None, title=None, normalize=True):\n",
        "    \"\"\"Imshow for Tensor.\"\"\"\n",
        "    if ax is None:\n",
        "        fig, ax = plt.subplots()\n",
        "    image = image.numpy().transpose((1, 2, 0))\n",
        "\n",
        "    if normalize:\n",
        "        mean = np.array([0.485, 0.456, 0.406])\n",
        "        std = np.array([0.229, 0.224, 0.225])\n",
        "        image = std * image + mean\n",
        "        image = np.clip(image, 0, 1)\n",
        "\n",
        "    ax.imshow(image)\n",
        "    ax.spines['top'].set_visible(False)\n",
        "    ax.spines['right'].set_visible(False)\n",
        "    ax.spines['left'].set_visible(False)\n",
        "    ax.spines['bottom'].set_visible(False)\n",
        "    ax.tick_params(axis='both', length=0)\n",
        "    ax.set_xticklabels('')\n",
        "    ax.set_yticklabels('')\n",
        "    return ax\n",
        "\n",
        "\n",
        "print(\" Sizes of Datasets: \", len(valid_dataset), len(train_dataset))\n",
        "# Displaying Training Images\n",
        "images, labels = next(iter(dataloaders['train']))\n",
        "\n",
        "fig, axes = plt.subplots(figsize=(16,5), ncols=5)\n",
        "for ii in range(5):\n",
        "    ax = axes[ii]\n",
        "    #ax.set_title(label_map[class_names[labels[ii].item()]])\n",
        "    imshow(images[ii], ax=ax, normalize=True)\n",
        "\n",
        "       "
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            " Sizes of Datasets:  814 7330\n",
            " Image Size torch.Size([32, 3, 224, 224])\n",
            "tensor(38)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA48AAACsCAYAAADbuZkKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 1152x360 with 5 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VAsRUjg73LTl",
        "colab_type": "text"
      },
      "source": [
        "### Building and Training The Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "0ZIjfhXs3LTm",
        "colab_type": "code",
        "outputId": "e36a3030-370b-4a02-d495-003b4f5a9a77",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 67
        }
      },
      "source": [
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "\n",
        "model_name = 'inception' #vgg\n",
        "if model_name == 'densenet':\n",
        "    model = models.densenet161(pretrained=True)\n",
        "    num_in_features = 2208\n",
        "    print(model)\n",
        "elif model_name == 'vgg':\n",
        "    model = models.vgg19(pretrained=True)\n",
        "    num_in_features = 25088\n",
        "    print(model.classifier)\n",
        "elif model_name == 'resnet':\n",
        "    model = models.resnet152(pretrained=True) \n",
        "    #model = models.resnet34(pretrained=True)\n",
        "    num_in_features = 2048 #512 \n",
        "    print(model.fc)\n",
        "elif model_name == 'inception':\n",
        "  model = models.inception_v3(pretrained=True)\n",
        "  model.aux_logits=False\n",
        "  num_in_features = 2048\n",
        "  print(model.fc)\n",
        "else:\n",
        "    print(\"Unknown model, please choose 'densenet' or 'vgg'\")"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading: \"https://download.pytorch.org/models/resnet152-b121ed2d.pth\" to /root/.cache/torch/checkpoints/resnet152-b121ed2d.pth\n",
            "100%|██████████| 241530880/241530880 [00:02<00:00, 111005169.64it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Linear(in_features=2048, out_features=1000, bias=True)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "LCc1gmnR3LTp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Freezing model parameters and defining the fully connected network to be attached to the model, loss function and the optimizer.\n",
        "#We there after put the model on the GPUs\n",
        "for param in model.parameters():\n",
        "  param.require_grad = False\n",
        "  \n",
        "# Create Custom Classifier\n",
        "def build_classifier(num_in_features, hidden_layers, num_out_features):\n",
        "   \n",
        "    classifier = nn.Sequential()\n",
        "    if hidden_layers == None:\n",
        "      \n",
        "        classifier.add_module('fc0', nn.Linear(num_in_features, 196))\n",
        "        \n",
        "    else:\n",
        "      \n",
        "        layer_sizes = zip(hidden_layers[:-1], hidden_layers[1:])\n",
        "        classifier.add_module('fc0', nn.Linear(num_in_features, hidden_layers[0]))\n",
        "        classifier.add_module('relu0', nn.ReLU())\n",
        "        classifier.add_module('drop0', nn.Dropout(.6))\n",
        "        \n",
        "#         classifier.add_module('relu1', nn.ReLU())\n",
        "#         classifier.add_module('drop1', nn.Dropout(.5))\n",
        "        \n",
        "        for i, (h1, h2) in enumerate(layer_sizes):\n",
        "            classifier.add_module('fc'+str(i+1), nn.Linear(h1, h2))\n",
        "            classifier.add_module('relu'+str(i+1), nn.ReLU())\n",
        "            classifier.add_module('drop'+str(i+1), nn.Dropout(.5))\n",
        "        classifier.add_module('output', nn.Linear(hidden_layers[-1], num_out_features))\n",
        "        \n",
        "    return classifier"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "V4W-jtIF3LTr",
        "colab_type": "code",
        "outputId": "5805f733-a70f-46a5-cfbf-e72b35c114d0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 67
        }
      },
      "source": [
        "hidden_layers = None #[1050 , 500] #[4096, 1024] #None#[4096, 1024, 256][512, 256, 128] [1050 , 500]\n",
        "\n",
        "classifier = build_classifier(num_in_features, hidden_layers, 196)\n",
        "print(classifier)\n",
        "\n",
        " # Defining model hyperparameters\n",
        "if model_name == 'densenet':\n",
        "    model.classifier = classifier\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    optimizer = optim.Adadelta(model.parameters()) # Adadelta #weight optim.Adam(model.parameters(), lr=0.001, momentum=0.9)\n",
        "    #optimizer_conv = optim.SGD(model.parameters(), lr=0.0001, weight_decay=0.001, momentum=0.9)\n",
        "   # Decay LR by a factor of 0.1 every 4 epochs\n",
        "    sched = optim.lr_scheduler.StepLR(optimizer, step_size=4)\n",
        "elif model_name == 'vgg':\n",
        "    model.classifier = classifier\n",
        "    #criterion = nn.NLLLoss()\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    #optimizer = optim.SGD(model.parameters(), lr=0.0001,weight_decay=0.001, momentum=0.9)\n",
        "    optimizer = optim.Adam(model.classifier.parameters(), lr=0.0001)\n",
        "    sched = optim.lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.1)\n",
        "elif model_name == 'resnet':\n",
        "  model.fc = classifier\n",
        "  criterion = nn.CrossEntropyLoss()\n",
        "  optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)\n",
        "  sched = optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='max', patience=3, threshold = 0.9)\n",
        "  #sched = optim.lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.1)\n",
        "\n",
        "#     #criterion = nn.NLLLoss()\n",
        "#     optimizer = optim.Adam(model.fc.parameters(), lr= 0.00001)\n",
        "#     sched = optim.lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.1)\n",
        "#criterion = nn.CrossEntropyLoss()\n",
        "\n",
        "elif model_name == 'inception':\n",
        "    model.fc = classifier\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    #optimizer = optim.Adadelta(model.parameters()) # Adadelta #weight optim.Adam(model.parameters(), lr=0.001, momentum=0.9)\n",
        "    optimizer = optim.SGD(model.parameters(), lr=0.001,momentum=0.9)\n",
        "    #sched = optim.lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.1)\n",
        "else:\n",
        "    pass"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Sequential(\n",
            "  (fc0): Linear(in_features=2048, out_features=196, bias=True)\n",
            ")\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XVVK3cQ63LTv",
        "colab_type": "text"
      },
      "source": [
        "#### Training the Model\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "3q0BjJYK3LTv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def train_model(model, criterion, optimizer, sched, num_epochs=5,device='cuda'):\n",
        "    start = time.time()\n",
        "    train_results = []\n",
        "    valid_results = []\n",
        "    best_model_wts = copy.deepcopy(model.state_dict())\n",
        "    best_acc = 0.0\n",
        "\n",
        "    for epoch in range(num_epochs):\n",
        "        \n",
        "        print('Epoch {}/{}'.format(epoch+1, num_epochs))\n",
        "        \n",
        "        print('-' * 10)\n",
        "        # Each epoch has a training and validation phase\n",
        "        for phase in ['train', 'valid']:\n",
        "            if phase == 'train':            \n",
        "              model.train()  # Set model to training mode\n",
        "            else:\n",
        "              model.eval()   # Set model to evaluate mode\n",
        "\n",
        "            running_loss = 0.0\n",
        "            running_corrects = 0\n",
        "\n",
        "            # Iterate over data.\n",
        "            for inputs, labels in dataloaders[phase]:\n",
        "                inputs = inputs.to(device)\n",
        "                labels = labels.to(device)\n",
        "\n",
        "                # zero the parameter gradients\n",
        "                optimizer.zero_grad()\n",
        "\n",
        "                # forward\n",
        "                # track history if only in train\n",
        "                with torch.set_grad_enabled(phase == 'train'):\n",
        "                    outputs = model(inputs)\n",
        "                    _, preds = torch.max(outputs, 1)\n",
        "                    loss = criterion(outputs, labels)\n",
        "\n",
        "                    # backward + optimize only if in training phase\n",
        "                    if phase == 'train':\n",
        "                        #sched.step()\n",
        "                        loss.backward()\n",
        "                        \n",
        "                        optimizer.step()\n",
        "\n",
        "                # statistics\n",
        "                running_loss += loss.item() * inputs.size(0)\n",
        "                running_corrects += torch.sum(preds == labels.data)\n",
        "\n",
        "            epoch_loss = running_loss / dataset_sizes[phase]\n",
        "            epoch_acc = running_corrects.double() / dataset_sizes[phase]\n",
        "            \n",
        "            # calculate average time over an epoch\n",
        "            #elapshed_epoch = time.time() - start/            \n",
        "            #print('Epoch {}/{} - completed in: {:.0f}m {:.0f}s'.format(epoch+1, num_epochs,elapshed_epoch // 60, elapshed_epoch % 60))\n",
        "            \n",
        "            if(phase == 'train'):\n",
        "              train_results.append([epoch_loss,epoch_acc])\n",
        "            if(phase == 'valid'):\n",
        "              #sched.step(epoch_acc)\n",
        "              valid_results.append([epoch_loss,epoch_acc])\n",
        "                                   \n",
        "            print('{} Loss: {:.4f} Acc: {:.4f}'.format(phase, epoch_loss, epoch_acc))\n",
        "\n",
        "            # deep copy the model (Early Stopping) and Saving our model, when we get best accuracy\n",
        "            if phase == 'valid' and epoch_acc > best_acc:\n",
        "                best_acc = epoch_acc\n",
        "                best_model_wts = copy.deepcopy(model.state_dict())       \n",
        "                #print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(valid_loss_min,valid_loss))\n",
        "                #model_save_name = \"ResNetDeepFlowers.pt\"\n",
        "                model_save_name = \"resnetCars2.pt\"\n",
        "                path = F\"/content/drive/My Drive/{model_save_name}\"\n",
        "                torch.save(model.state_dict(), path)        \n",
        "\n",
        "        print()\n",
        "\n",
        "    time_elapsed = time.time() - start\n",
        "    print('Training complete in {:.0f}m {:.0f}s'.format(\n",
        "        time_elapsed // 60, time_elapsed % 60))\n",
        "    print('Best val Acc: {:4f}'.format(best_acc))\n",
        "\n",
        "    #load best model weights\n",
        "    model.load_state_dict(best_model_wts)\n",
        "    \n",
        "    return model,train_results,valid_results\n",
        "  #Resnet34 = 68.8%, 50 epochs, vggDeepFlowers"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LpIHPq2DUfin",
        "colab_type": "code",
        "outputId": "06d7a13d-a8db-4632-a26b-aec8ebb929c7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "model.load_state_dict(torch.load('/content/drive/My Drive/resnetCars1.pt'))\n",
        "model.to(device)"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ResNet(\n",
              "  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
              "  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "  (relu): ReLU(inplace)\n",
              "  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
              "  (layer1): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (layer2): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
              "        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (3): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (4): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (5): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (6): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (7): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (layer3): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
              "        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (3): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (4): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (5): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (6): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (7): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (8): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (9): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (10): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (11): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (12): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (13): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (14): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (15): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (16): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (17): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (18): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (19): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (20): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (21): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (22): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (23): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (24): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (25): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (26): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (27): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (28): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (29): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (30): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (31): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (32): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (33): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (34): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (35): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (layer4): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
              "        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
              "  (fc): Sequential(\n",
              "    (fc0): Linear(in_features=2048, out_features=196, bias=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "LWqzxWTq3LTy",
        "colab_type": "code",
        "outputId": "33692abd-4095-456f-d0ba-44894e514650",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "epochs = 60\n",
        "model.to(device)\n",
        "model,train_results,valid_results = train_model(model, criterion, optimizer, sched, epochs)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/60\n",
            "----------\n",
            "train Loss: 0.5672 Acc: 0.8441\n",
            "valid Loss: 0.6750 Acc: 0.8329\n",
            "\n",
            "Epoch 2/60\n",
            "----------\n",
            "train Loss: 0.6184 Acc: 0.8357\n",
            "valid Loss: 0.5980 Acc: 0.8415\n",
            "\n",
            "Epoch 3/60\n",
            "----------\n",
            "train Loss: 0.5695 Acc: 0.8487\n",
            "valid Loss: 0.5503 Acc: 0.8575\n",
            "\n",
            "Epoch 4/60\n",
            "----------\n",
            "train Loss: 0.6020 Acc: 0.8342\n",
            "valid Loss: 0.6641 Acc: 0.8403\n",
            "\n",
            "Epoch 5/60\n",
            "----------\n",
            "train Loss: 0.5546 Acc: 0.8498\n",
            "valid Loss: 0.6638 Acc: 0.8219\n",
            "\n",
            "Epoch 6/60\n",
            "----------\n",
            "train Loss: 0.5482 Acc: 0.8528\n",
            "valid Loss: 0.6211 Acc: 0.8329\n",
            "\n",
            "Epoch 7/60\n",
            "----------\n",
            "train Loss: 0.5467 Acc: 0.8540\n",
            "valid Loss: 0.6727 Acc: 0.8292\n",
            "\n",
            "Epoch 8/60\n",
            "----------\n",
            "train Loss: 0.5692 Acc: 0.8460\n",
            "valid Loss: 0.6882 Acc: 0.8206\n",
            "\n",
            "Epoch 9/60\n",
            "----------\n",
            "train Loss: 0.5403 Acc: 0.8539\n",
            "valid Loss: 0.6068 Acc: 0.8280\n",
            "\n",
            "Epoch 10/60\n",
            "----------\n",
            "train Loss: 0.5232 Acc: 0.8613\n",
            "valid Loss: 0.6875 Acc: 0.8256\n",
            "\n",
            "Epoch 11/60\n",
            "----------\n",
            "train Loss: 0.5515 Acc: 0.8569\n",
            "valid Loss: 0.7818 Acc: 0.8010\n",
            "\n",
            "Epoch 12/60\n",
            "----------\n",
            "train Loss: 0.5198 Acc: 0.8644\n",
            "valid Loss: 0.5460 Acc: 0.8514\n",
            "\n",
            "Epoch 13/60\n",
            "----------\n",
            "train Loss: 0.5003 Acc: 0.8636\n",
            "valid Loss: 0.5965 Acc: 0.8452\n",
            "\n",
            "Epoch 14/60\n",
            "----------\n",
            "train Loss: 0.4900 Acc: 0.8726\n",
            "valid Loss: 0.6383 Acc: 0.8256\n",
            "\n",
            "Epoch 15/60\n",
            "----------\n",
            "train Loss: 0.4948 Acc: 0.8681\n",
            "valid Loss: 0.5916 Acc: 0.8354\n",
            "\n",
            "Epoch 16/60\n",
            "----------\n",
            "train Loss: 0.4240 Acc: 0.8888\n",
            "valid Loss: 0.6380 Acc: 0.8317\n",
            "\n",
            "Epoch 17/60\n",
            "----------\n",
            "train Loss: 0.4751 Acc: 0.8701\n",
            "valid Loss: 0.6738 Acc: 0.8182\n",
            "\n",
            "Epoch 18/60\n",
            "----------\n",
            "train Loss: 0.4970 Acc: 0.8701\n",
            "valid Loss: 0.7092 Acc: 0.8268\n",
            "\n",
            "Epoch 19/60\n",
            "----------\n",
            "train Loss: 0.4799 Acc: 0.8722\n",
            "valid Loss: 0.5697 Acc: 0.8563\n",
            "\n",
            "Epoch 20/60\n",
            "----------\n",
            "train Loss: 0.4565 Acc: 0.8769\n",
            "valid Loss: 0.6195 Acc: 0.8280\n",
            "\n",
            "Epoch 21/60\n",
            "----------\n",
            "train Loss: 0.4943 Acc: 0.8726\n",
            "valid Loss: 0.5910 Acc: 0.8378\n",
            "\n",
            "Epoch 22/60\n",
            "----------\n",
            "train Loss: 0.4524 Acc: 0.8825\n",
            "valid Loss: 0.6588 Acc: 0.8354\n",
            "\n",
            "Epoch 23/60\n",
            "----------\n",
            "train Loss: 0.4407 Acc: 0.8808\n",
            "valid Loss: 0.6355 Acc: 0.8391\n",
            "\n",
            "Epoch 24/60\n",
            "----------\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-17-c5184f54306a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mepochs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m60\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdevice\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mtrain_results\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mvalid_results\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcriterion\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msched\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m<ipython-input-15-65c04ab9beea>\u001b[0m in \u001b[0;36mtrain_model\u001b[0;34m(model, criterion, optimizer, sched, num_epochs, device)\u001b[0m\n\u001b[1;32m     39\u001b[0m                     \u001b[0;32mif\u001b[0m \u001b[0mphase\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'train'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     40\u001b[0m                         \u001b[0;31m#sched.step()\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 41\u001b[0;31m                         \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     42\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m                         \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torch/tensor.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(self, gradient, retain_graph, create_graph)\u001b[0m\n\u001b[1;32m    105\u001b[0m                 \u001b[0mproducts\u001b[0m\u001b[0;34m.\u001b[0m \u001b[0mDefaults\u001b[0m \u001b[0mto\u001b[0m\u001b[0;31m \u001b[0m\u001b[0;31m`\u001b[0m\u001b[0;31m`\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;31m`\u001b[0m\u001b[0;31m`\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    106\u001b[0m         \"\"\"\n\u001b[0;32m--> 107\u001b[0;31m         \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mautograd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgradient\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    108\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    109\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mregister_hook\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhook\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/torch/autograd/__init__.py\u001b[0m in \u001b[0;36mbackward\u001b[0;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables)\u001b[0m\n\u001b[1;32m     91\u001b[0m     Variable._execution_engine.run_backward(\n\u001b[1;32m     92\u001b[0m         \u001b[0mtensors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mgrad_tensors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mretain_graph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcreate_graph\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 93\u001b[0;31m         allow_unreachable=True)  # allow_unreachable flag\n\u001b[0m\u001b[1;32m     94\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     95\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "jTE0-kf43LT1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# resnet34\n",
        "# criterion = nn.CrossEntropyLoss()\n",
        "# optimizer = optim.SGD(model_ft.parameters(), lr=0.01, momentum=0.9)\n",
        "\n",
        "# \"\"\"\n",
        "# probably not the best metric to track, but we are tracking the training accuracy and measuring whether\n",
        "# it increases by atleast 0.9 per epoch and if it hasn't increased by 0.9 reduce the lr by 0.1x.\n",
        "# However in this model it did not benefit me.\n",
        "# \"\"\"\n",
        "# lrscheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='max', patience=3, threshold = 0.9)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JscSfTWpx08c",
        "colab_type": "text"
      },
      "source": [
        "### Load Trained Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SQ2Y5ZQqxzu6",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "ad2b2931-7f3b-4ec3-cce2-ac344221553a"
      },
      "source": [
        "#Load trained model and continue training\n",
        "model.load_state_dict(torch.load('/content/drive/My Drive/resnetCars2.pt'))\n",
        "model.to(device)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ResNet(\n",
              "  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n",
              "  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "  (relu): ReLU(inplace)\n",
              "  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)\n",
              "  (layer1): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (layer2): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
              "        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (3): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (4): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (5): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (6): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (7): Bottleneck(\n",
              "      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (layer3): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
              "        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (3): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (4): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (5): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (6): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (7): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (8): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (9): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (10): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (11): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (12): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (13): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (14): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (15): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (16): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (17): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (18): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (19): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (20): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (21): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (22): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (23): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (24): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (25): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (26): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (27): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (28): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (29): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (30): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (31): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (32): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (33): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (34): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (35): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (layer4): Sequential(\n",
              "    (0): Bottleneck(\n",
              "      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "      (downsample): Sequential(\n",
              "        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)\n",
              "        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      )\n",
              "    )\n",
              "    (1): Bottleneck(\n",
              "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "    (2): Bottleneck(\n",
              "      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (relu): ReLU(inplace)\n",
              "    )\n",
              "  )\n",
              "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
              "  (fc): Sequential(\n",
              "    (fc0): Linear(in_features=2048, out_features=196, bias=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OQBdiYnpyT40",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UvjW1BgxyT8g",
        "colab_type": "text"
      },
      "source": [
        "### Predictions On Test Sample"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O7-zTuwXx62m",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from pathlib import Path\n",
        "\n",
        "\n",
        "# Load the datasets with ImageFolder\n",
        "# data_dir = 'flower_data'\n",
        "# train_dir = data_dir + '/train'\n",
        "# valid_dir = data_dir + '/valid'\n",
        "# test_dir = data_dir +'/test set'\n",
        "# test_dirs = data_dir +'/test/1'\n",
        "label_df = pd.read_csv('names.csv', names=[\"label\"])\n",
        "\n",
        "test_dir = 'car_data/test'\n",
        "\n",
        "with torch.no_grad():\n",
        "  model.eval()\n",
        "  # test_data = next(iter(dataloaders['test']))\n",
        "  \n",
        "  dataset = datasets.ImageFolder(test_dir,transform=data_transforms['test'])\n",
        "  \n",
        "  testloader = torch.utils.data.DataLoader(dataset, batch_size=64,\n",
        "                                             shuffle=False, num_workers=2)\n",
        " \n",
        "  image_names = []\n",
        "  for index in testloader.dataset.imgs:\n",
        "    #image_names.append(index[0].split('/')[-1])\n",
        "    image_names.append(Path(index[0]).stem)\n",
        "\n",
        "  results = []\n",
        "  #results.append(image_names)\n",
        "  \n",
        "  for inputs, labels in testloader:\n",
        "    inputs = inputs.to(device)\n",
        "    labels = labels.to(device)\n",
        "    outputs = model(inputs)\n",
        "    _, predicted = torch.max(outputs, 1)\n",
        "    for i in predicted:\n",
        "        results.append(int(i)+1)\n",
        "        "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "N0RjgoS6Tmbb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "c256b893-181c-4217-da9a-2cef6fe896ea"
      },
      "source": [
        "print(\"Predictions on Test Set:\")\n",
        "df = pd.DataFrame({'Id': image_names, 'Predicted': results})\n",
        "pd.set_option('display.max_colwidth', -1)\n",
        "\n",
        "# df = df.sort_values(by=['Id']) \n",
        "df.to_csv('/content/drive/My Drive/predictions.csv')\n",
        "df"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Predictions on Test Set:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Id</th>\n",
              "      <th>Predicted</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>00076</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>00457</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>00684</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>01117</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>01167</td>\n",
              "      <td>125</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>01199</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>01226</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>01461</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>01518</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>01538</td>\n",
              "      <td>149</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>01802</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>01887</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>01918</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>01946</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>02017</td>\n",
              "      <td>149</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>02205</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>02217</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>02267</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>02547</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>02602</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>02665</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>02835</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>03073</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>03094</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>03193</td>\n",
              "      <td>149</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>03246</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>03585</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>03640</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>03876</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>03904</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8011</th>\n",
              "      <td>01736</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8012</th>\n",
              "      <td>01778</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8013</th>\n",
              "      <td>02007</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8014</th>\n",
              "      <td>02493</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8015</th>\n",
              "      <td>02985</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8016</th>\n",
              "      <td>03157</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8017</th>\n",
              "      <td>03397</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8018</th>\n",
              "      <td>03483</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8019</th>\n",
              "      <td>03590</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8020</th>\n",
              "      <td>03711</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8021</th>\n",
              "      <td>03742</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8022</th>\n",
              "      <td>03908</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8023</th>\n",
              "      <td>04189</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8024</th>\n",
              "      <td>04307</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8025</th>\n",
              "      <td>04490</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8026</th>\n",
              "      <td>04973</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8027</th>\n",
              "      <td>05236</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8028</th>\n",
              "      <td>05253</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8029</th>\n",
              "      <td>05706</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8030</th>\n",
              "      <td>05723</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8031</th>\n",
              "      <td>05728</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8032</th>\n",
              "      <td>05789</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8033</th>\n",
              "      <td>05966</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8034</th>\n",
              "      <td>06028</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8035</th>\n",
              "      <td>06479</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8036</th>\n",
              "      <td>06615</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8037</th>\n",
              "      <td>06886</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8038</th>\n",
              "      <td>07236</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8039</th>\n",
              "      <td>07467</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8040</th>\n",
              "      <td>07592</td>\n",
              "      <td>196</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>8041 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         Id  Predicted\n",
              "0     00076  1        \n",
              "1     00457  1        \n",
              "2     00684  1        \n",
              "3     01117  1        \n",
              "4     01167  125      \n",
              "5     01199  1        \n",
              "6     01226  1        \n",
              "7     01461  108      \n",
              "8     01518  1        \n",
              "9     01538  149      \n",
              "10    01802  1        \n",
              "11    01887  1        \n",
              "12    01918  1        \n",
              "13    01946  1        \n",
              "14    02017  149      \n",
              "15    02205  1        \n",
              "16    02217  1        \n",
              "17    02267  1        \n",
              "18    02547  1        \n",
              "19    02602  1        \n",
              "20    02665  1        \n",
              "21    02835  1        \n",
              "22    03073  1        \n",
              "23    03094  1        \n",
              "24    03193  149      \n",
              "25    03246  1        \n",
              "26    03585  1        \n",
              "27    03640  1        \n",
              "28    03876  1        \n",
              "29    03904  1        \n",
              "...     ... ..        \n",
              "8011  01736  196      \n",
              "8012  01778  196      \n",
              "8013  02007  196      \n",
              "8014  02493  196      \n",
              "8015  02985  196      \n",
              "8016  03157  196      \n",
              "8017  03397  196      \n",
              "8018  03483  196      \n",
              "8019  03590  196      \n",
              "8020  03711  196      \n",
              "8021  03742  196      \n",
              "8022  03908  196      \n",
              "8023  04189  196      \n",
              "8024  04307  196      \n",
              "8025  04490  196      \n",
              "8026  04973  196      \n",
              "8027  05236  196      \n",
              "8028  05253  196      \n",
              "8029  05706  196      \n",
              "8030  05723  196      \n",
              "8031  05728  196      \n",
              "8032  05789  196      \n",
              "8033  05966  196      \n",
              "8034  06028  196      \n",
              "8035  06479  196      \n",
              "8036  06615  196      \n",
              "8037  06886  196      \n",
              "8038  07236  196      \n",
              "8039  07467  196      \n",
              "8040  07592  196      \n",
              "\n",
              "[8041 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x_uLjZ54yeuY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}