terrainRecModel.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "29514af8-7429-470a-a6c8-f4b82974544a",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import numpy as np\n",
    "import torch\n",
    "from torch import nn\n",
    "from torch import optim\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets, transforms, models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e603c93c-9bcb-4133-833b-228c7c246dae",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['grass', 'marshy', 'rocky', 'sandy']\n"
     ]
    }
   ],
   "source": [
    "data_dir = 'Data'\n",
    "\n",
    "def load_split_train_test(datadir, valid_size = .2):\n",
    "    train_transforms = transforms.Compose([transforms.RandomRotation(30), \n",
    "                                       transforms.RandomResizedCrop(224), \n",
    "                                       transforms.RandomHorizontalFlip(),\n",
    "                                       transforms.Resize([224, 224]),\n",
    "                                       transforms.ToTensor(),\n",
    "                                       transforms.Normalize([0.485, 0.456, 0.406], \n",
    "                                                            [0.229, 0.224, 0.225]) \n",
    "                                       ])\n",
    "\n",
    "    test_transforms = transforms.Compose([transforms.Resize([224, 224]),\n",
    "                                      transforms.ToTensor(),\n",
    "                                      transforms.Normalize([0.485, 0.456, 0.406],\n",
    "                                                           [0.229, 0.224, 0.225])\n",
    "                                      ])\n",
    "\n",
    "    train_data = datasets.ImageFolder(datadir, transform=train_transforms)\n",
    "    test_data = datasets.ImageFolder(datadir, transform=test_transforms)\n",
    "\n",
    "    num_train = len(train_data)\n",
    "    indices = list(range(num_train))\n",
    "    split = int(np.floor(valid_size * num_train))\n",
    "    np.random.shuffle(indices)\n",
    "    from torch.utils.data.sampler import SubsetRandomSampler\n",
    "    train_idx, test_idx = indices[split:], indices[:split]\n",
    "    train_sampler = SubsetRandomSampler(train_idx)\n",
    "    test_sampler = SubsetRandomSampler(test_idx)\n",
    "    trainloader = torch.utils.data.DataLoader(train_data, sampler=train_sampler, batch_size=64)\n",
    "    testloader = torch.utils.data.DataLoader(test_data, sampler=test_sampler, batch_size=64)\n",
    "    return trainloader, testloader\n",
    "\n",
    "trainloader, testloader = load_split_train_test(data_dir, .2)\n",
    "print(trainloader.dataset.classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "55350107-81d7-4583-a31e-9e5e00010d19",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/abhishekshandilya/pytorch-local/env/lib/python3.8/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "/Users/abhishekshandilya/pytorch-local/env/lib/python3.8/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet50_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet50_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n"
     ]
    },
    {
     "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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\n",
       "    )\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
       "  (fc): Linear(in_features=2048, out_features=1000, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "model = models.resnet50(pretrained=True)\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3e6ff914-11a0-45aa-b5a0-54625bb4371f",
   "metadata": {},
   "outputs": [
    {
     "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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\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=True)\n",
       "    )\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n",
       "  (fc): Sequential(\n",
       "    (0): Linear(in_features=2048, out_features=512, bias=True)\n",
       "    (1): ReLU()\n",
       "    (2): Dropout(p=0.2, inplace=False)\n",
       "    (3): Linear(in_features=512, out_features=10, bias=True)\n",
       "    (4): LogSoftmax(dim=1)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Freeze parameters so we don't backprop through them\n",
    "for param in model.parameters():\n",
    "    param.requires_grad = False\n",
    "    \n",
    "model.fc = nn.Sequential(nn.Linear(2048, 512),\n",
    "                                 nn.ReLU(),\n",
    "                                 nn.Dropout(0.2),\n",
    "                                 nn.Linear(512, 10),\n",
    "                                 nn.LogSoftmax(dim=1))\n",
    "criterion = nn.NLLLoss()\n",
    "optimizer = optim.Adam(model.fc.parameters(), lr=0.003)\n",
    "model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a7bbb9a5-9d38-4075-923c-4641a2dbf930",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10.. Train loss: 2.718.. Test loss: 1.624.. Test accuracy: 0.560\n",
      "Epoch 1/10.. Train loss: 1.399.. Test loss: 0.728.. Test accuracy: 0.702\n",
      "Epoch 2/10.. Train loss: 0.942.. Test loss: 0.676.. Test accuracy: 0.692\n",
      "Epoch 2/10.. Train loss: 0.721.. Test loss: 0.496.. Test accuracy: 0.819\n",
      "Epoch 3/10.. Train loss: 0.567.. Test loss: 0.470.. Test accuracy: 0.853\n",
      "Epoch 3/10.. Train loss: 0.550.. Test loss: 0.380.. Test accuracy: 0.891\n",
      "Epoch 4/10.. Train loss: 0.523.. Test loss: 0.393.. Test accuracy: 0.862\n",
      "Epoch 4/10.. Train loss: 0.627.. Test loss: 0.450.. Test accuracy: 0.808\n",
      "Epoch 5/10.. Train loss: 0.496.. Test loss: 0.468.. Test accuracy: 0.811\n",
      "Epoch 5/10.. Train loss: 0.502.. Test loss: 0.498.. Test accuracy: 0.818\n",
      "Epoch 5/10.. Train loss: 0.478.. Test loss: 0.397.. Test accuracy: 0.868\n",
      "Epoch 6/10.. Train loss: 0.527.. Test loss: 0.427.. Test accuracy: 0.852\n",
      "Epoch 6/10.. Train loss: 0.456.. Test loss: 0.338.. Test accuracy: 0.865\n",
      "Epoch 7/10.. Train loss: 0.515.. Test loss: 0.368.. Test accuracy: 0.865\n",
      "Epoch 7/10.. Train loss: 0.564.. Test loss: 0.364.. Test accuracy: 0.860\n",
      "Epoch 8/10.. Train loss: 0.499.. Test loss: 0.405.. Test accuracy: 0.844\n",
      "Epoch 8/10.. Train loss: 0.464.. Test loss: 0.345.. Test accuracy: 0.881\n",
      "Epoch 9/10.. Train loss: 0.520.. Test loss: 0.358.. Test accuracy: 0.876\n",
      "Epoch 9/10.. Train loss: 0.462.. Test loss: 0.350.. Test accuracy: 0.873\n",
      "Epoch 10/10.. Train loss: 0.550.. Test loss: 0.395.. Test accuracy: 0.878\n",
      "Epoch 10/10.. Train loss: 0.494.. Test loss: 0.567.. Test accuracy: 0.772\n",
      "Epoch 10/10.. Train loss: 0.557.. Test loss: 0.337.. Test accuracy: 0.883\n"
     ]
    }
   ],
   "source": [
    "epochs = 10\n",
    "steps = 0\n",
    "running_loss = 0\n",
    "print_every = 10\n",
    "train_losses, test_losses = [], []\n",
    "\n",
    "for epoch in range(epochs):\n",
    "    for inputs, labels in trainloader:\n",
    "        steps += 1\n",
    "        inputs, labels = inputs.to(device), labels.to(device)\n",
    "        optimizer.zero_grad()\n",
    "        logps = model.forward(inputs)\n",
    "        loss = criterion(logps, labels)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        running_loss += loss.item()\n",
    "        \n",
    "        if steps % print_every == 0:\n",
    "            test_loss = 0\n",
    "            accuracy = 0\n",
    "            model.eval()\n",
    "            with torch.no_grad():\n",
    "                for inputs, labels in testloader:\n",
    "                    inputs, labels = inputs.to(device), labels.to(device)\n",
    "                    logps = model.forward(inputs)\n",
    "                    batch_loss = criterion(logps, labels)\n",
    "                    test_loss += batch_loss.item()\n",
    "                    \n",
    "                    ps = torch.exp(logps)\n",
    "                    top_p, top_class = ps.topk(1, dim=1)\n",
    "                    equals = top_class == labels.view(*top_class.shape)\n",
    "                    accuracy += torch.mean(equals.type(torch.FloatTensor)).item()\n",
    "\n",
    "            train_losses.append(running_loss/len(trainloader))\n",
    "            test_losses.append(test_loss/len(testloader))                    \n",
    "            print(f\"Epoch {epoch+1}/{epochs}.. \"\n",
    "                  f\"Train loss: {running_loss/print_every:.3f}.. \"\n",
    "                  f\"Test loss: {test_loss/len(testloader):.3f}.. \"\n",
    "                  f\"Test accuracy: {accuracy/len(testloader):.3f}\")\n",
    "            running_loss = 0\n",
    "            model.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "090b7449-780a-42e2-9602-e04de56c91a9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 413,
       "width": 547
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(train_losses, label='Training loss')\n",
    "plt.plot(test_losses, label='Validation loss')\n",
    "plt.legend(frameon=False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "923121c6-cf83-4fa5-8e53-e3a199359306",
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.save(model, 'model/aerialDelter.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "939905d3-0565-420f-9658-e4aaeaa8e919",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.17"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}