FlowerImageClassifierpytorch.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "FlowerImageClassifierpytorch.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/adnan119/udacity_aipwpnd_projects/blob/master/FlowerImageClassifierpytorch.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "CBpKMkK_YsA4",
        "colab_type": "code",
        "outputId": "c440810f-73f9-4b3a-f0e9-5137e45e66de",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "cell_type": "code",
      "source": [
        "# http://pytorch.org/\n",
        "from os.path import exists\n",
        "from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag\n",
        "platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())\n",
        "cuda_output = !ldconfig -p|grep cudart.so|sed -e 's/.*\\.\\([0-9]*\\)\\.\\([0-9]*\\)$/cu\\1\\2/'\n",
        "accelerator = cuda_output[0] if exists('/dev/nvidia0') else 'cpu'\n",
        "\n",
        "!pip install -q http://download.pytorch.org/whl/{accelerator}/torch-0.4.1-{platform}-linux_x86_64.whl torchvision\n",
        "import torch"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tcmalloc: large alloc 1073750016 bytes == 0x58444000 @  0x7f904dbf82a4 0x591a07 0x5b5d56 0x502e9a 0x506859 0x502209 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x507641 0x502209 0x502f3d 0x506859 0x504c28 0x502540 0x502f3d 0x507641 0x504c28 0x502540 0x502f3d 0x507641\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "qhVkgODIYl7q",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Imports here\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import torch\n",
        "import numpy as np\n",
        "from torch import nn\n",
        "from torch import optim\n",
        "import torch.nn.functional as F\n",
        "from torchvision import datasets, transforms, models\n",
        "from torch.optim import lr_scheduler\n",
        "from PIL import Image\n",
        "%matplotlib inline\n",
        "from matplotlib.ticker import FormatStrFormatter\n",
        "import json\n",
        "from collections import OrderedDict\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "D_8_ehQjsNQa",
        "colab_type": "code",
        "outputId": "960fffbc-57e0-4fd8-9ef7-fb9f98e47d6f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "#checking for gpu\n",
        "\n",
        "train_on_gpu = torch.cuda.is_available()\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": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "CUDA is available!  Training on GPU ...\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Lh9AlJXCZCfw",
        "colab_type": "code",
        "outputId": "3b7aabd9-2d66-4f10-90d7-0dfda05ffaf6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 69
        }
      },
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')\n",
        "!ls\n",
        "!pwd"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n",
            "gdrive\tsample_data\n",
            "/content\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "H1SgXih14Rxa",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "data_dir = 'gdrive/My Drive/ProjectWorks/flower_data'\n",
        "train_dir = data_dir + '/train'\n",
        "valid_dir = data_dir + '/valid'\n",
        "test_dir = data_dir + '/test'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "m_UmPaj55Qon",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "train_transform = transforms.Compose([transforms.RandomRotation(30),\n",
        "                                       transforms.RandomResizedCrop(224),\n",
        "                                       transforms.RandomHorizontalFlip(),\n",
        "                                       transforms.RandomVerticalFlip(),\n",
        "                                       transforms.RandomRotation(90),\n",
        "                                       transforms.ToTensor(),\n",
        "                                       transforms.Normalize([0.485, 0.456, 0.406], \n",
        "                                                            [0.229, 0.224, 0.225])])\n",
        "\n",
        "val_transform = transforms.Compose([transforms.Resize(256),\n",
        "                                      transforms.CenterCrop(224),\n",
        "                                      transforms.ToTensor(),\n",
        "                                      transforms.Normalize([0.485, 0.456, 0.406], \n",
        "                                                           [0.229, 0.224, 0.225])])\n",
        "\n",
        "test_transform = transforms.Compose([transforms.Resize(256),\n",
        "                                      transforms.CenterCrop(224),\n",
        "                                      transforms.ToTensor(),\n",
        "                                     transforms.Normalize([0.485, 0.456, 0.406], \n",
        "                                                           [0.229, 0.224, 0.225])])\n",
        "\n",
        "train_data = datasets.ImageFolder(train_dir, transform=train_transform)\n",
        "val_data = datasets.ImageFolder(valid_dir, transform = val_transform)\n",
        "test_data = datasets.ImageFolder(test_dir, transform = test_transform)\n",
        "\n",
        "trainloader = torch.utils.data.DataLoader(train_data, batch_size=32, shuffle=True)\n",
        "validloader = torch.utils.data.DataLoader(val_data, batch_size=16)\n",
        "testloader = torch.utils.data.DataLoader(test_data, batch_size=16)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Q8mo1wxd7pJ9",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "with open('gdrive/My Drive/ProjectWorks/cat_to_name.json', 'r') as f:\n",
        "    cat_to_name = json.load(f)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "tz1gRNsn8QdN",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = models.resnet152(pretrained=True)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "znNb_T6u_Cz8",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "for param in model.parameters():\n",
        "    param.requires_grad = False"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "2HvuTvweAq1L",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "classifier = nn.Sequential(OrderedDict([\n",
        "                          #('adaptive_pool',nn.AdaptiveAvgPool2d((1,1))),\n",
        "                          #('adaptive_maxpool',nn.AdaptiveMaxPool2d((1,1))),\n",
        "                          #('flatten',nn.Flatten()),\n",
        "                          ('batch_norm',nn.BatchNorm1d(2048,eps=1e-05, momentum=0.1, affine=True)),\n",
        "                          ('Dropout',nn.Dropout(0)),\n",
        "                          ('fc1', nn.Linear(2048, 520)),\n",
        "                          ('activation', nn.ReLU()),\n",
        "                          ('fc2', nn.Linear(520, 102)),\n",
        "                          ('output', nn.LogSoftmax(dim=1))\n",
        "                          ]))\n",
        "model.fc = classifier"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "2I8AlkvSBMTW",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "if train_on_gpu:\n",
        "  model.cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "dwiZ-iUImqca",
        "colab_type": "code",
        "outputId": "d8401a11-7149-4113-8193-39dcd1655896",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 8509
        }
      },
      "cell_type": "code",
      "source": [
        "model"
      ],
      "execution_count": 0,
      "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): AvgPool2d(kernel_size=7, stride=1, padding=0)\n",
              "  (fc): Sequential(\n",
              "    (batch_norm): BatchNorm1d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (Dropout): Dropout(p=0)\n",
              "    (fc1): Linear(in_features=2048, out_features=520, bias=True)\n",
              "    (activation): ReLU()\n",
              "    (fc2): Linear(in_features=520, out_features=102, bias=True)\n",
              "    (output): LogSoftmax()\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "metadata": {
        "id": "Xp17F7bzEbz6",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "criterion = nn.NLLLoss()\n",
        "optimizer = optim.Adam(model.fc.parameters(), lr=0.0003)\n",
        "lrsheduler = lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Ui4z1Nv4GzzB",
        "colab_type": "code",
        "outputId": "3bdd07de-9923-4d26-901f-030c48822ed7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 729
        }
      },
      "cell_type": "code",
      "source": [
        "epochs = 30\n",
        "print_every = 40\n",
        "steps = 0\n",
        "valid_loss_min = np.Inf\n",
        "total = 0\n",
        "correct = 0\n",
        "\n",
        "# change to cuda\n",
        "model.to('cuda')\n",
        "validation_loss = 0\n",
        "for e in range(epochs):\n",
        "    training_loss = 0\n",
        "    lrsheduler.step(validation_loss)\n",
        "    model.train()\n",
        "    for ii, (inputs, labels) in enumerate(trainloader):\n",
        "        steps += 1\n",
        "        \n",
        "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
        "        \n",
        "        optimizer.zero_grad()\n",
        "        \n",
        "        # Forward and backward passes\n",
        "        outputs = model.forward(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "        \n",
        "        training_loss += loss.item()*inputs.size(0)\n",
        "        \n",
        "        \n",
        "    model.eval()\n",
        "    for ii, (inputs, labels) in enumerate(validloader):\n",
        "\n",
        "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
        "        outputs = model.forward(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "\n",
        "        validation_loss += loss.item()*inputs.size(0)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total += labels.size(0)\n",
        "        correct += (predicted == labels).sum().item()\n",
        "        \n",
        "    training_loss = training_loss/len(trainloader.dataset)\n",
        "    validation_loss = validation_loss/len(validloader.dataset)\n",
        "    _correct = (correct*100)/total\n",
        "    \n",
        "    \n",
        "    \n",
        "    print('Epoch: {} \\tTraining Loss: {:.6f} \\tValidation Loss: {:.6f} \\tAccuracy: {:.6f}'.format(\n",
        "        e, training_loss, validation_loss, _correct))\n",
        "    \n",
        "    if validation_loss <= valid_loss_min:\n",
        "        print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(\n",
        "        valid_loss_min,\n",
        "        validation_loss))\n",
        "        \n",
        "    #to save the check-points\n",
        "        model.class_to_idx = train_data.class_to_idx\n",
        "        checkpoint = {'architecture': 'densenet161',\n",
        "                'classifier': model.fc,\n",
        "                'class_to_idx': model.class_to_idx,\n",
        "                'state_dict': model.state_dict()}\n",
        "        \n",
        "        torch.save(checkpoint, 'model_challenge_2.pt')\n",
        "        valid_loss_min = validation_loss"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 0 \tTraining Loss: 1.539038 \tValidation Loss: 0.744360 \tAccuracy: 81.907090\n",
            "Validation loss decreased (inf --> 0.744360).  Saving model ...\n",
            "Epoch: 1 \tTraining Loss: 0.881517 \tValidation Loss: 0.573746 \tAccuracy: 83.007335\n",
            "Validation loss decreased (0.744360 --> 0.573746).  Saving model ...\n",
            "Epoch: 2 \tTraining Loss: 0.733855 \tValidation Loss: 0.539933 \tAccuracy: 83.414833\n",
            "Validation loss decreased (0.573746 --> 0.539933).  Saving model ...\n",
            "Epoch: 3 \tTraining Loss: 0.679763 \tValidation Loss: 0.416090 \tAccuracy: 84.688264\n",
            "Validation loss decreased (0.539933 --> 0.416090).  Saving model ...\n",
            "Epoch: 4 \tTraining Loss: 0.626644 \tValidation Loss: 0.420032 \tAccuracy: 85.330073\n",
            "Epoch: 5 \tTraining Loss: 0.604332 \tValidation Loss: 0.432946 \tAccuracy: 85.656072\n",
            "Epoch: 6 \tTraining Loss: 0.575961 \tValidation Loss: 0.385054 \tAccuracy: 86.150891\n",
            "Validation loss decreased (0.416090 --> 0.385054).  Saving model ...\n",
            "Epoch: 7 \tTraining Loss: 0.536464 \tValidation Loss: 0.340811 \tAccuracy: 86.751222\n",
            "Validation loss decreased (0.385054 --> 0.340811).  Saving model ...\n",
            "Epoch: 8 \tTraining Loss: 0.516592 \tValidation Loss: 0.326602 \tAccuracy: 87.299647\n",
            "Validation loss decreased (0.340811 --> 0.326602).  Saving model ...\n",
            "Epoch: 9 \tTraining Loss: 0.479925 \tValidation Loss: 0.342817 \tAccuracy: 87.701711\n",
            "Epoch: 10 \tTraining Loss: 0.479037 \tValidation Loss: 0.330411 \tAccuracy: 88.019560\n",
            "Epoch: 11 \tTraining Loss: 0.473605 \tValidation Loss: 0.298588 \tAccuracy: 88.386308\n",
            "Validation loss decreased (0.326602 --> 0.298588).  Saving model ...\n",
            "Epoch: 12 \tTraining Loss: 0.461381 \tValidation Loss: 0.306965 \tAccuracy: 88.621403\n",
            "Epoch: 13 \tTraining Loss: 0.446777 \tValidation Loss: 0.318102 \tAccuracy: 88.918966\n",
            "Epoch: 14 \tTraining Loss: 0.440515 \tValidation Loss: 0.299002 \tAccuracy: 89.152404\n",
            "Epoch: 15 \tTraining Loss: 0.406658 \tValidation Loss: 0.321269 \tAccuracy: 89.349022\n",
            "Epoch: 16 \tTraining Loss: 0.415199 \tValidation Loss: 0.291841 \tAccuracy: 89.544082\n",
            "Validation loss decreased (0.298588 --> 0.291841).  Saving model ...\n",
            "Epoch: 17 \tTraining Loss: 0.422467 \tValidation Loss: 0.316751 \tAccuracy: 89.663135\n",
            "Epoch: 18 \tTraining Loss: 0.410164 \tValidation Loss: 0.302335 \tAccuracy: 89.827564\n",
            "Epoch: 19 \tTraining Loss: 0.386495 \tValidation Loss: 0.288787 \tAccuracy: 89.987775\n",
            "Validation loss decreased (0.291841 --> 0.288787).  Saving model ...\n",
            "Epoch: 20 \tTraining Loss: 0.370301 \tValidation Loss: 0.329418 \tAccuracy: 90.086157\n",
            "Epoch: 21 \tTraining Loss: 0.392593 \tValidation Loss: 0.314035 \tAccuracy: 90.175595\n",
            "Epoch: 22 \tTraining Loss: 0.385011 \tValidation Loss: 0.292594 \tAccuracy: 90.267886\n",
            "Epoch: 23 \tTraining Loss: 0.371019 \tValidation Loss: 0.296980 \tAccuracy: 90.367767\n",
            "Epoch: 24 \tTraining Loss: 0.374025 \tValidation Loss: 0.283758 \tAccuracy: 90.479218\n",
            "Validation loss decreased (0.288787 --> 0.283758).  Saving model ...\n",
            "Epoch: 25 \tTraining Loss: 0.376613 \tValidation Loss: 0.293581 \tAccuracy: 90.572691\n",
            "Epoch: 26 \tTraining Loss: 0.362201 \tValidation Loss: 0.294001 \tAccuracy: 90.668297\n",
            "Epoch: 27 \tTraining Loss: 0.352116 \tValidation Loss: 0.313372 \tAccuracy: 90.726511\n",
            "Epoch: 28 \tTraining Loss: 0.347365 \tValidation Loss: 0.295471 \tAccuracy: 90.822865\n",
            "Epoch: 29 \tTraining Loss: 0.328240 \tValidation Loss: 0.300211 \tAccuracy: 90.892421\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "co8qb8w3HVhv",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "for param in model.parameters():\n",
        "        param.requires_grad = True\n",
        "\n",
        "optimizer = optim.Adagrad(model.parameters(), lr=0.0003)\n",
        "lrsheduler1 = lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.1)\n",
        "lrsheduler2 = lr_scheduler.ReduceLROnPlateau(optimizer, mode = 'min', factor = 0.1, patience = 3)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "2LLd0xYqI9Vn",
        "colab_type": "code",
        "outputId": "8b8dfb12-a788-465b-a2d5-9c348fe100db",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 660
        }
      },
      "cell_type": "code",
      "source": [
        "epochs = 30\n",
        "print_every = 40\n",
        "steps = 0\n",
        "valid_loss_min = validation_loss\n",
        "validation_loss = 0\n",
        "total = 0\n",
        "correct = 0\n",
        "\n",
        "# change to cuda\n",
        "model.to('cuda')\n",
        "lrsheduler = lrsheduler1\n",
        "for e in range(epochs):\n",
        "    training_loss = 0\n",
        "    #validation_loss = 0\n",
        "    lrsheduler.step(validation_loss)\n",
        "    model.train()\n",
        "    for ii, (inputs, labels) in enumerate(trainloader):\n",
        "        steps += 1\n",
        "        \n",
        "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
        "        \n",
        "        optimizer.zero_grad()\n",
        "        \n",
        "        # Forward and backward passes\n",
        "        outputs = model.forward(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "        \n",
        "        training_loss += loss.item()*inputs.size(0)\n",
        "            \n",
        "    model.eval()\n",
        "    for ii, (inputs, labels) in enumerate(validloader):\n",
        "\n",
        "        inputs, labels = inputs.to('cuda'), labels.to('cuda')\n",
        "        outputs = model.forward(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "\n",
        "        validation_loss += loss.item()*inputs.size(0)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total += labels.size(0)\n",
        "        correct += (predicted == labels).sum().item()\n",
        "        \n",
        "    training_loss = training_loss/len(trainloader.dataset)\n",
        "    validation_loss = validation_loss/len(validloader.dataset)\n",
        "    _correct = (correct*100)/total\n",
        "    \n",
        "    if epochs > 14 and _correct < 97:\n",
        "      lrsheduler = lrsheduler2\n",
        "      \n",
        "    print('Epoch: {} \\tTraining Loss: {:.6f} \\tValidation Loss: {:.6f} \\tAccuracy: {:.6f}'.format(\n",
        "        e, training_loss, validation_loss, _correct))\n",
        "    \n",
        "    if validation_loss <= valid_loss_min:\n",
        "        print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(\n",
        "        valid_loss_min,\n",
        "        validation_loss))\n",
        "        \n",
        "    #to save the check-points\n",
        "        model.class_to_idx = train_data.class_to_idx\n",
        "        checkpoint = {'architecture': 'densenet161',\n",
        "                'classifier': model.fc,\n",
        "                'class_to_idx': model.class_to_idx,\n",
        "                'state_dict': model.state_dict()}\n",
        "        \n",
        "        torch.save(checkpoint, 'model_challenge_2.pt')\n",
        "        valid_loss_min = validation_loss"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 0 \tTraining Loss: 0.361540 \tValidation Loss: 0.183394 \tAccuracy: 95.965770\n",
            "Validation loss decreased (0.300211 --> 0.183394).  Saving model ...\n",
            "Epoch: 1 \tTraining Loss: 0.221729 \tValidation Loss: 0.162280 \tAccuracy: 96.332518\n",
            "Validation loss decreased (0.183394 --> 0.162280).  Saving model ...\n",
            "Epoch: 2 \tTraining Loss: 0.187998 \tValidation Loss: 0.154184 \tAccuracy: 96.291769\n",
            "Validation loss decreased (0.162280 --> 0.154184).  Saving model ...\n",
            "Epoch: 3 \tTraining Loss: 0.161436 \tValidation Loss: 0.138445 \tAccuracy: 96.454768\n",
            "Validation loss decreased (0.154184 --> 0.138445).  Saving model ...\n",
            "Epoch: 4 \tTraining Loss: 0.150172 \tValidation Loss: 0.124756 \tAccuracy: 96.699267\n",
            "Validation loss decreased (0.138445 --> 0.124756).  Saving model ...\n",
            "Epoch: 5 \tTraining Loss: 0.125097 \tValidation Loss: 0.132811 \tAccuracy: 96.862266\n",
            "Epoch: 6 \tTraining Loss: 0.144379 \tValidation Loss: 0.134275 \tAccuracy: 96.961229\n",
            "Epoch: 7 \tTraining Loss: 0.123217 \tValidation Loss: 0.137105 \tAccuracy: 96.989609\n",
            "Epoch: 8 \tTraining Loss: 0.137088 \tValidation Loss: 0.129119 \tAccuracy: 97.066015\n",
            "Epoch: 9 \tTraining Loss: 0.129136 \tValidation Loss: 0.125491 \tAccuracy: 97.114914\n",
            "Epoch: 10 \tTraining Loss: 0.121535 \tValidation Loss: 0.127329 \tAccuracy: 97.177150\n",
            "Epoch: 11 \tTraining Loss: 0.105662 \tValidation Loss: 0.127889 \tAccuracy: 97.208639\n",
            "Epoch: 12 \tTraining Loss: 0.118091 \tValidation Loss: 0.122246 \tAccuracy: 97.244687\n",
            "Validation loss decreased (0.124756 --> 0.122246).  Saving model ...\n",
            "Epoch: 13 \tTraining Loss: 0.118573 \tValidation Loss: 0.126165 \tAccuracy: 97.258121\n",
            "Epoch: 14 \tTraining Loss: 0.109232 \tValidation Loss: 0.119163 \tAccuracy: 97.277914\n",
            "Validation loss decreased (0.122246 --> 0.119163).  Saving model ...\n",
            "Epoch: 15 \tTraining Loss: 0.111997 \tValidation Loss: 0.121702 \tAccuracy: 97.302873\n",
            "Epoch: 16 \tTraining Loss: 0.121840 \tValidation Loss: 0.124508 \tAccuracy: 97.310513\n",
            "Epoch: 17 \tTraining Loss: 0.119623 \tValidation Loss: 0.128956 \tAccuracy: 97.317305\n",
            "Epoch: 18 \tTraining Loss: 0.105860 \tValidation Loss: 0.121932 \tAccuracy: 97.342684\n",
            "Epoch: 19 \tTraining Loss: 0.106254 \tValidation Loss: 0.119484 \tAccuracy: 97.371638\n",
            "Epoch: 20 \tTraining Loss: 0.112381 \tValidation Loss: 0.124700 \tAccuracy: 97.386192\n",
            "Epoch: 21 \tTraining Loss: 0.104575 \tValidation Loss: 0.126338 \tAccuracy: 97.399422\n",
            "Epoch: 22 \tTraining Loss: 0.111677 \tValidation Loss: 0.123419 \tAccuracy: 97.411502\n",
            "Epoch: 23 \tTraining Loss: 0.116116 \tValidation Loss: 0.122212 \tAccuracy: 97.432763\n",
            "Epoch: 24 \tTraining Loss: 0.108180 \tValidation Loss: 0.121977 \tAccuracy: 97.437653\n",
            "Epoch: 25 \tTraining Loss: 0.109987 \tValidation Loss: 0.122526 \tAccuracy: 97.446869\n",
            "Epoch: 26 \tTraining Loss: 0.114956 \tValidation Loss: 0.123712 \tAccuracy: 97.455402\n",
            "Epoch: 27 \tTraining Loss: 0.098646 \tValidation Loss: 0.123834 \tAccuracy: 97.463325\n",
            "Epoch: 28 \tTraining Loss: 0.107355 \tValidation Loss: 0.123436 \tAccuracy: 97.462271\n",
            "Epoch: 29 \tTraining Loss: 0.111772 \tValidation Loss: 0.123570 \tAccuracy: 97.465363\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "9RnjUNpQJHyO",
        "colab_type": "code",
        "outputId": "69f45fc9-e3c2-4a17-ade2-5b7c0f600ae6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "# TODO: Do validation on the test set\n",
        "def check_accuracy_on_test(testloader):    \n",
        "    correct = 0\n",
        "    total = 0\n",
        "    model.to('cuda:0')\n",
        "    with torch.no_grad():\n",
        "        for data in testloader:\n",
        "            images, labels = data\n",
        "            images, labels = images.to('cuda'), labels.to('cuda')\n",
        "            outputs = model(images)\n",
        "            _, predicted = torch.max(outputs.data, 1)\n",
        "            total += labels.size(0)\n",
        "            correct += (predicted == labels).sum().item()\n",
        "\n",
        "    print('Accuracy of the network on the test images: %d %%' % (100 * correct / total))\n",
        "    \n",
        "check_accuracy_on_test(testloader)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy of the network on the test images: 96 %\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Jr8Sg_tVLHBG",
        "colab_type": "code",
        "outputId": "cf0268ea-749e-4e51-ec54-6fa740a42b67",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 8509
        }
      },
      "cell_type": "code",
      "source": [
        "# TODO: Write a function that loads a checkpoint and rebuilds the model\n",
        "def load_checkpoint():\n",
        "    \"\"\"\n",
        "    Loads deep learning model checkpoint.\n",
        "    \"\"\"\n",
        "    \n",
        "    # Load the saved file\n",
        "    checkpoint = torch.load(\"model_challenge_2.pt\")\n",
        "    \n",
        "    # Freeze parameters so we don't backprop through them\n",
        "    for param in model.parameters(): param.requires_grad = False\n",
        "    \n",
        "    # Load stuff from checkpoint\n",
        "    model.class_to_idx = checkpoint['class_to_idx']\n",
        "    model.fc = checkpoint['classifier']\n",
        "    model.load_state_dict(checkpoint['state_dict'])\n",
        "\n",
        "    \n",
        "    return model\n",
        "\n",
        "load_checkpoint()"
      ],
      "execution_count": 0,
      "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): AvgPool2d(kernel_size=7, stride=1, padding=0)\n",
              "  (fc): Sequential(\n",
              "    (batch_norm): BatchNorm1d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
              "    (Dropout): Dropout(p=0)\n",
              "    (fc1): Linear(in_features=2048, out_features=520, bias=True)\n",
              "    (activation): ReLU()\n",
              "    (fc2): Linear(in_features=520, out_features=102, bias=True)\n",
              "    (output): LogSoftmax()\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "metadata": {
        "id": "aE_Ey5d3Iaat",
        "colab_type": "code",
        "outputId": "ac9f14f3-d104-4c1e-f305-fcb55c66ca45",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "def process_image(image):\n",
        "    img_pil = Image.open(image)\n",
        "   \n",
        "    image_process = transforms.Compose([\n",
        "        transforms.Resize(256),\n",
        "        transforms.CenterCrop(224),\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])\n",
        "    ])\n",
        "    \n",
        "    img_tensor = image_process(img_pil)\n",
        "    \n",
        "    return img_tensor\n",
        "    \n",
        "    \n",
        "    # TODO: Process a PIL image for use in a PyTorch model\n",
        "\n",
        "img = (data_dir + '/test' + '/53/' + 'image_03672.jpg')\n",
        "img = process_image(img)\n",
        "print(img.shape)\n",
        "\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([3, 224, 224])\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "kBB_Hg2nKI3_",
        "colab_type": "code",
        "outputId": "09740d87-f0b9-41e1-c46b-49e10878bc1b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        }
      },
      "cell_type": "code",
      "source": [
        "def imshow(image, ax=None, title=None):\n",
        "    if ax is None:\n",
        "        fig, ax = plt.subplots()\n",
        "    \n",
        "    # PyTorch tensors assume the color channel is the first dimension\n",
        "    # but matplotlib assumes is the third dimension\n",
        "    image = image.numpy().transpose((1, 2, 0))\n",
        "    \n",
        "    # Undo preprocessing\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",
        "    \n",
        "    # Image needs to be clipped between 0 and 1 or it looks like noise when displayed\n",
        "    image = np.clip(image, 0, 1)\n",
        "    \n",
        "    ax.imshow(image)\n",
        "    \n",
        "    return ax\n",
        "\n",
        "imshow(process_image(data_dir + '/test' + '/53/' + 'image_03672.jpg'))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f19c7341710>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 63
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "B295tPlEKPvN",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def predict(image_path, model, top_k=5):\n",
        "    ''' Predict the class (or classes) of an image using a trained deep learning model.\n",
        "    \n",
        "    image_path: string. Path to image, directly to image and not to folder.\n",
        "    model: pytorch neural network.\n",
        "    top_k: integer. The top K classes to be calculated\n",
        "    \n",
        "    returns top_probabilities(k), top_labels\n",
        "    '''\n",
        "    \n",
        "    # No need for GPU on this part (just causes problems)\n",
        "    model.to(\"cpu\")\n",
        "    \n",
        "    # Set model to evaluate\n",
        "    model.eval();\n",
        "\n",
        "    # Convert image from numpy to torch\n",
        "    torch_image = torch.from_numpy(np.expand_dims(process_image(image_path), \n",
        "                                                  axis=0)).type(torch.FloatTensor).to(\"cpu\")\n",
        "\n",
        "    # Find probabilities (results) by passing through the function (note the log softmax means that its on a log scale)\n",
        "    log_probs = model.forward(torch_image)\n",
        "\n",
        "    # Convert to linear scale\n",
        "    linear_probs = torch.exp(log_probs)\n",
        "\n",
        "    # Find the top 5 results\n",
        "    top_probs, top_labels = linear_probs.topk(top_k)\n",
        "    \n",
        "    # Detatch all of the details\n",
        "    top_probs = np.array(top_probs.detach())[0] # This is not the correct way to do it but the correct way isnt working thanks to cpu/gpu issues so I don't care.\n",
        "    top_labels = np.array(top_labels.detach())[0]\n",
        "    \n",
        "    # Convert to classes\n",
        "    idx_to_class = {val: key for key, val in    \n",
        "                                      model.class_to_idx.items()}\n",
        "    top_labels = [idx_to_class[lab] for lab in top_labels]\n",
        "    top_flowers = [cat_to_name[lab] for lab in top_labels]\n",
        "    \n",
        "    return top_probs, top_labels, top_flowers"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Ka5jNzyEL-j_",
        "colab_type": "code",
        "outputId": "f4907277-d2ac-4b56-c7d8-8dc4e462af84",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 630
        }
      },
      "cell_type": "code",
      "source": [
        "image_path = data_dir + '/test' + '/53/' + 'image_03672.jpg'\n",
        "\n",
        "# Set up plot\n",
        "plt.figure(figsize = (6,10))\n",
        "ax = plt.subplot(2,1,1)\n",
        "\n",
        "# Set up title\n",
        "flower_num = image_path.split('/')[5]\n",
        "title_ = cat_to_name[flower_num]\n",
        "\n",
        "# Plot flower\n",
        "img = process_image(image_path)\n",
        "imshow(img, ax, title = title_);\n",
        "\n",
        "# Make prediction\n",
        "probs, labs, flowers = predict(image_path, model) \n",
        "\n",
        "# Plot bar chart\n",
        "plt.subplot(2,1,2)\n",
        "sns.barplot(x=probs, y=flowers, color=sns.color_palette()[0]);\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py:1428: FutureWarning: remove_na is deprecated and is a private function. Do not use.\n",
            "  stat_data = remove_na(group_data)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAawAAAJCCAYAAACcbvPDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nP6NkXJ7qabA6AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "Psdo9g1IMLPa",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}