{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Deep Learning Models -- A collection of various deep learning architectures, models, and tips for TensorFlow and PyTorch in Jupyter Notebooks.\n",
    "- Author: Sebastian Raschka\n",
    "- GitHub Repository: https://github.com/rasbt/deeplearning-models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Transfer Learning Example (VGG16 pre-trained on ImageNet for Cifar-10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sebastian Raschka \n",
      "\n",
      "CPython 3.7.3\n",
      "IPython 7.9.0\n",
      "\n",
      "torch 1.3.0\n",
      "torchvision 0.4.1a0+d94043a\n"
     ]
    }
   ],
   "source": [
    "%load_ext watermark\n",
    "%watermark -a 'Sebastian Raschka' -v -p torch,torchvision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import time\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torchvision import datasets\n",
    "from torchvision import transforms\n",
    "from torch.utils.data import DataLoader\n",
    "\n",
    "#######################################\n",
    "### PRE-TRAINED MODELS AVAILABLE HERE\n",
    "## https://pytorch.org/docs/stable/torchvision/models.html\n",
    "from torchvision import models\n",
    "#######################################\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    torch.backends.cudnn.deterministic = True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading an Example Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this example, we are going to work with CIFAR-10, because it's small (smaller) than ImageNet and fast to downliad. However, note that in a \"real-world application\", images with dimension > 224x224 are recommended when working with modelsthat have been trained on ImageNet images with > 224x224 size. Here, we resize the images as a workaround.\n",
    "\n",
    "- Note that due to the average pooling in the final layer, it is also possible to feed in 32x32-pixel images directly. However, I noticed that the performance is rather low (~65% test accuracy after 10 and 100 epochs).\n",
    "\n",
    "- Also note that we we normalize the images with the following parameters\n",
    "\n",
    "```\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406],\n",
    "                          std=[0.229, 0.224, 0.225])\n",
    "```\n",
    "\n",
    "which have been used for training the model originally on ImageNet."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Device: cuda:0\n",
      "Files already downloaded and verified\n",
      "Image batch dimensions: torch.Size([128, 3, 224, 224])\n",
      "Image label dimensions: torch.Size([128])\n"
     ]
    }
   ],
   "source": [
    "##########################\n",
    "### SETTINGS\n",
    "##########################\n",
    "\n",
    "# Device\n",
    "DEVICE = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "print('Device:', DEVICE)\n",
    "\n",
    "NUM_CLASSES = 10\n",
    "\n",
    "# Hyperparameters\n",
    "random_seed = 1\n",
    "learning_rate = 0.0001\n",
    "num_epochs = 10\n",
    "batch_size = 128\n",
    "\n",
    "\n",
    "##########################\n",
    "### MNIST DATASET\n",
    "##########################\n",
    "\n",
    "custom_transform = transforms.Compose([\n",
    "    transforms.Resize((224, 224)),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize(mean=[0.485, 0.456, 0.406],\n",
    "                          std=[0.229, 0.224, 0.225])\n",
    "])\n",
    "\n",
    "## Note that this particular normalization scheme is\n",
    "## necessary since it was used for pre-training\n",
    "## the network on ImageNet.\n",
    "## These are the channel-means and standard deviations\n",
    "## for z-score normalization.\n",
    "\n",
    "\n",
    "train_dataset = datasets.CIFAR10(root='data', \n",
    "                                 train=True, \n",
    "                                 transform=custom_transform,\n",
    "                                 download=True)\n",
    "\n",
    "test_dataset = datasets.CIFAR10(root='data', \n",
    "                                train=False, \n",
    "                                transform=custom_transform)\n",
    "\n",
    "\n",
    "train_loader = DataLoader(dataset=train_dataset, \n",
    "                          batch_size=batch_size, \n",
    "                          num_workers=8,\n",
    "                          shuffle=True)\n",
    "\n",
    "test_loader = DataLoader(dataset=test_dataset, \n",
    "                         batch_size=batch_size, \n",
    "                         num_workers=8,\n",
    "                         shuffle=False)\n",
    "\n",
    "# Checking the dataset\n",
    "for images, labels in train_loader:  \n",
    "    print('Image batch dimensions:', images.shape)\n",
    "    print('Image label dimensions:', labels.shape)\n",
    "    break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the Pre-Trained Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, we are going to use VGG16 as an example for transfer learning from `torchvision`. A list of all pre-trained models is available at\n",
    "\n",
    "- https://pytorch.org/docs/stable/torchvision/models.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "VGG(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (1): ReLU(inplace=True)\n",
       "    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (3): ReLU(inplace=True)\n",
       "    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (6): ReLU(inplace=True)\n",
       "    (7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (8): ReLU(inplace=True)\n",
       "    (9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (11): ReLU(inplace=True)\n",
       "    (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (13): ReLU(inplace=True)\n",
       "    (14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (15): ReLU(inplace=True)\n",
       "    (16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (18): ReLU(inplace=True)\n",
       "    (19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (20): ReLU(inplace=True)\n",
       "    (21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (22): ReLU(inplace=True)\n",
       "    (23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "    (24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (25): ReLU(inplace=True)\n",
       "    (26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (27): ReLU(inplace=True)\n",
       "    (28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (29): ReLU(inplace=True)\n",
       "    (30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
       "  )\n",
       "  (avgpool): AdaptiveAvgPool2d(output_size=(7, 7))\n",
       "  (classifier): Sequential(\n",
       "    (0): Linear(in_features=25088, out_features=4096, bias=True)\n",
       "    (1): ReLU(inplace=True)\n",
       "    (2): Dropout(p=0.5, inplace=False)\n",
       "    (3): Linear(in_features=4096, out_features=4096, bias=True)\n",
       "    (4): ReLU(inplace=True)\n",
       "    (5): Dropout(p=0.5, inplace=False)\n",
       "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = models.vgg16(pretrained=True)\n",
    "model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Freezing the Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we are going to freeze the whole model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "for param in model.parameters():\n",
    "    param.requires_grad = False"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, Assume we want to train the penultimate layer (here, `model.classifier[3]` as we can see from the model structure above, which I am pasting as a reference below:\n",
    "\n",
    "```\n",
    "VGG(\n",
    "  (features): Sequential(\n",
    "    (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
    "    (1): ReLU(inplace=True)\n",
    "    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
    "...\n",
    "    (29): ReLU(inplace=True)\n",
    "    (30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
    "  )\n",
    "  (avgpool): AdaptiveAvgPool2d(output_size=(7, 7))\n",
    "  (classifier): Sequential(\n",
    "    (0): Linear(in_features=25088, out_features=4096, bias=True)\n",
    "    (1): ReLU(inplace=True)\n",
    "    (2): Dropout(p=0.5, inplace=False)\n",
    "->    (3): Linear(in_features=4096, out_features=4096, bias=True)\n",
    "    (4): ReLU(inplace=True)\n",
    "    (5): Dropout(p=0.5, inplace=False)\n",
    "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
    "  )\n",
    ")\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.classifier[3].requires_grad = True"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, replace the output layer with your own output layer (here, we actually add two more output layers):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.classifier[6] = nn.Sequential(\n",
    "                      nn.Linear(4096, 512), \n",
    "                      nn.ReLU(), \n",
    "                      nn.Dropout(0.5),\n",
    "                      nn.Linear(512, NUM_CLASSES))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training (as usual)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = model.to(DEVICE)\n",
    "optimizer = torch.optim.Adam(model.parameters())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 001/010 | Batch 0000/0391 | Cost: 2.3803\n",
      "Epoch: 001/010 | Batch 0050/0391 | Cost: 0.8682\n",
      "Epoch: 001/010 | Batch 0100/0391 | Cost: 0.7765\n",
      "Epoch: 001/010 | Batch 0150/0391 | Cost: 0.5888\n",
      "Epoch: 001/010 | Batch 0200/0391 | Cost: 0.6363\n",
      "Epoch: 001/010 | Batch 0250/0391 | Cost: 0.5042\n",
      "Epoch: 001/010 | Batch 0300/0391 | Cost: 0.7212\n",
      "Epoch: 001/010 | Batch 0350/0391 | Cost: 0.5531\n",
      "Epoch: 001/010 | Train: 83.768% | Loss: 0.470\n",
      "Time elapsed: 10.51 min\n",
      "Epoch: 002/010 | Batch 0000/0391 | Cost: 0.4309\n",
      "Epoch: 002/010 | Batch 0050/0391 | Cost: 0.5423\n",
      "Epoch: 002/010 | Batch 0100/0391 | Cost: 0.6057\n",
      "Epoch: 002/010 | Batch 0150/0391 | Cost: 0.7861\n",
      "Epoch: 002/010 | Batch 0200/0391 | Cost: 0.5859\n",
      "Epoch: 002/010 | Batch 0250/0391 | Cost: 0.6265\n",
      "Epoch: 002/010 | Batch 0300/0391 | Cost: 0.5713\n",
      "Epoch: 002/010 | Batch 0350/0391 | Cost: 0.4664\n",
      "Epoch: 002/010 | Train: 84.770% | Loss: 0.435\n",
      "Time elapsed: 21.05 min\n",
      "Epoch: 003/010 | Batch 0000/0391 | Cost: 0.5218\n",
      "Epoch: 003/010 | Batch 0050/0391 | Cost: 0.4995\n",
      "Epoch: 003/010 | Batch 0100/0391 | Cost: 0.5690\n",
      "Epoch: 003/010 | Batch 0150/0391 | Cost: 0.6084\n",
      "Epoch: 003/010 | Batch 0200/0391 | Cost: 0.6712\n",
      "Epoch: 003/010 | Batch 0250/0391 | Cost: 0.7230\n",
      "Epoch: 003/010 | Batch 0300/0391 | Cost: 0.6850\n",
      "Epoch: 003/010 | Batch 0350/0391 | Cost: 0.5648\n",
      "Epoch: 003/010 | Train: 85.626% | Loss: 0.418\n",
      "Time elapsed: 31.59 min\n",
      "Epoch: 004/010 | Batch 0000/0391 | Cost: 0.5770\n",
      "Epoch: 004/010 | Batch 0050/0391 | Cost: 0.5119\n",
      "Epoch: 004/010 | Batch 0100/0391 | Cost: 0.5196\n",
      "Epoch: 004/010 | Batch 0150/0391 | Cost: 0.6272\n",
      "Epoch: 004/010 | Batch 0200/0391 | Cost: 0.5175\n",
      "Epoch: 004/010 | Batch 0250/0391 | Cost: 0.5380\n",
      "Epoch: 004/010 | Batch 0300/0391 | Cost: 0.5041\n",
      "Epoch: 004/010 | Batch 0350/0391 | Cost: 0.6165\n",
      "Epoch: 004/010 | Train: 87.010% | Loss: 0.386\n",
      "Time elapsed: 42.13 min\n",
      "Epoch: 005/010 | Batch 0000/0391 | Cost: 0.6082\n",
      "Epoch: 005/010 | Batch 0050/0391 | Cost: 0.6508\n",
      "Epoch: 005/010 | Batch 0100/0391 | Cost: 0.5656\n",
      "Epoch: 005/010 | Batch 0150/0391 | Cost: 0.5483\n",
      "Epoch: 005/010 | Batch 0200/0391 | Cost: 0.5408\n",
      "Epoch: 005/010 | Batch 0250/0391 | Cost: 0.7091\n",
      "Epoch: 005/010 | Batch 0300/0391 | Cost: 0.5846\n",
      "Epoch: 005/010 | Batch 0350/0391 | Cost: 0.4931\n",
      "Epoch: 005/010 | Train: 87.088% | Loss: 0.372\n",
      "Time elapsed: 52.66 min\n",
      "Epoch: 006/010 | Batch 0000/0391 | Cost: 0.5629\n",
      "Epoch: 006/010 | Batch 0050/0391 | Cost: 0.4118\n",
      "Epoch: 006/010 | Batch 0100/0391 | Cost: 0.4184\n",
      "Epoch: 006/010 | Batch 0150/0391 | Cost: 0.5407\n",
      "Epoch: 006/010 | Batch 0200/0391 | Cost: 0.5839\n",
      "Epoch: 006/010 | Batch 0250/0391 | Cost: 0.5171\n",
      "Epoch: 006/010 | Batch 0300/0391 | Cost: 0.4679\n",
      "Epoch: 006/010 | Batch 0350/0391 | Cost: 0.5208\n",
      "Epoch: 006/010 | Train: 87.710% | Loss: 0.368\n",
      "Time elapsed: 63.20 min\n",
      "Epoch: 007/010 | Batch 0000/0391 | Cost: 0.4737\n",
      "Epoch: 007/010 | Batch 0050/0391 | Cost: 0.7670\n",
      "Epoch: 007/010 | Batch 0100/0391 | Cost: 0.4890\n",
      "Epoch: 007/010 | Batch 0150/0391 | Cost: 0.5645\n",
      "Epoch: 007/010 | Batch 0200/0391 | Cost: 0.6673\n",
      "Epoch: 007/010 | Batch 0250/0391 | Cost: 0.5325\n",
      "Epoch: 007/010 | Batch 0300/0391 | Cost: 0.6377\n",
      "Epoch: 007/010 | Batch 0350/0391 | Cost: 0.5301\n",
      "Epoch: 007/010 | Train: 87.692% | Loss: 0.354\n",
      "Time elapsed: 73.73 min\n",
      "Epoch: 008/010 | Batch 0000/0391 | Cost: 0.7276\n",
      "Epoch: 008/010 | Batch 0050/0391 | Cost: 0.5233\n",
      "Epoch: 008/010 | Batch 0100/0391 | Cost: 0.7512\n",
      "Epoch: 008/010 | Batch 0150/0391 | Cost: 0.5838\n",
      "Epoch: 008/010 | Batch 0200/0391 | Cost: 0.4164\n",
      "Epoch: 008/010 | Batch 0250/0391 | Cost: 0.6005\n",
      "Epoch: 008/010 | Batch 0300/0391 | Cost: 0.5340\n",
      "Epoch: 008/010 | Batch 0350/0391 | Cost: 0.4254\n",
      "Epoch: 008/010 | Train: 87.604% | Loss: 0.359\n",
      "Time elapsed: 84.28 min\n",
      "Epoch: 009/010 | Batch 0000/0391 | Cost: 0.7138\n",
      "Epoch: 009/010 | Batch 0050/0391 | Cost: 0.7279\n",
      "Epoch: 009/010 | Batch 0100/0391 | Cost: 0.3387\n",
      "Epoch: 009/010 | Batch 0150/0391 | Cost: 0.4552\n",
      "Epoch: 009/010 | Batch 0200/0391 | Cost: 0.3744\n",
      "Epoch: 009/010 | Batch 0250/0391 | Cost: 0.6198\n",
      "Epoch: 009/010 | Batch 0300/0391 | Cost: 0.5379\n",
      "Epoch: 009/010 | Batch 0350/0391 | Cost: 0.5648\n",
      "Epoch: 009/010 | Train: 88.338% | Loss: 0.341\n",
      "Time elapsed: 94.82 min\n",
      "Epoch: 010/010 | Batch 0000/0391 | Cost: 0.5407\n",
      "Epoch: 010/010 | Batch 0050/0391 | Cost: 0.4377\n",
      "Epoch: 010/010 | Batch 0100/0391 | Cost: 0.4832\n",
      "Epoch: 010/010 | Batch 0150/0391 | Cost: 0.4002\n",
      "Epoch: 010/010 | Batch 0200/0391 | Cost: 0.4990\n",
      "Epoch: 010/010 | Batch 0250/0391 | Cost: 0.3890\n",
      "Epoch: 010/010 | Batch 0300/0391 | Cost: 0.4749\n",
      "Epoch: 010/010 | Batch 0350/0391 | Cost: 0.7142\n",
      "Epoch: 010/010 | Train: 88.696% | Loss: 0.329\n",
      "Time elapsed: 105.35 min\n",
      "Total Training Time: 105.35 min\n"
     ]
    }
   ],
   "source": [
    "def compute_accuracy(model, data_loader):\n",
    "    model.eval()\n",
    "    correct_pred, num_examples = 0, 0\n",
    "    for i, (features, targets) in enumerate(data_loader):\n",
    "            \n",
    "        features = features.to(DEVICE)\n",
    "        targets = targets.to(DEVICE)\n",
    "\n",
    "        logits = model(features)\n",
    "        _, predicted_labels = torch.max(logits, 1)\n",
    "        num_examples += targets.size(0)\n",
    "        correct_pred += (predicted_labels == targets).sum()\n",
    "    return correct_pred.float()/num_examples * 100\n",
    "\n",
    "\n",
    "def compute_epoch_loss(model, data_loader):\n",
    "    model.eval()\n",
    "    curr_loss, num_examples = 0., 0\n",
    "    with torch.no_grad():\n",
    "        for features, targets in data_loader:\n",
    "            features = features.to(DEVICE)\n",
    "            targets = targets.to(DEVICE)\n",
    "            logits = model(features)\n",
    "            loss = F.cross_entropy(logits, targets, reduction='sum')\n",
    "            num_examples += targets.size(0)\n",
    "            curr_loss += loss\n",
    "\n",
    "        curr_loss = curr_loss / num_examples\n",
    "        return curr_loss\n",
    "    \n",
    "    \n",
    "\n",
    "start_time = time.time()\n",
    "for epoch in range(num_epochs):\n",
    "    \n",
    "    model.train()\n",
    "    for batch_idx, (features, targets) in enumerate(train_loader):\n",
    "        \n",
    "        features = features.to(DEVICE)\n",
    "        targets = targets.to(DEVICE)\n",
    "            \n",
    "        ### FORWARD AND BACK PROP\n",
    "        logits = model(features)\n",
    "        cost = F.cross_entropy(logits, targets)\n",
    "        optimizer.zero_grad()\n",
    "        \n",
    "        cost.backward()\n",
    "        \n",
    "        ### UPDATE MODEL PARAMETERS\n",
    "        optimizer.step()\n",
    "        \n",
    "        ### LOGGING\n",
    "        if not batch_idx % 50:\n",
    "            print ('Epoch: %03d/%03d | Batch %04d/%04d | Cost: %.4f' \n",
    "                   %(epoch+1, num_epochs, batch_idx, \n",
    "                     len(train_loader), cost))\n",
    "\n",
    "    model.eval()\n",
    "    with torch.set_grad_enabled(False): # save memory during inference\n",
    "        print('Epoch: %03d/%03d | Train: %.3f%% | Loss: %.3f' % (\n",
    "              epoch+1, num_epochs, \n",
    "              compute_accuracy(model, train_loader),\n",
    "              compute_epoch_loss(model, train_loader)))\n",
    "\n",
    "\n",
    "    print('Time elapsed: %.2f min' % ((time.time() - start_time)/60))\n",
    "    \n",
    "print('Total Training Time: %.2f min' % ((time.time() - start_time)/60))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test accuracy: 84.25%\n"
     ]
    }
   ],
   "source": [
    "with torch.set_grad_enabled(False): # save memory during inference\n",
    "    print('Test accuracy: %.2f%%' % (compute_accuracy(model, test_loader)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "classes = ('plane', 'car', 'bird', 'cat',\n",
    "           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')\n",
    "\n",
    "for batch_idx, (features, targets) in enumerate(test_loader):\n",
    "\n",
    "    features = features\n",
    "    targets = targets\n",
    "    break\n",
    "\n",
    "logits = model(features.to(DEVICE))\n",
    "_, predicted_labels = torch.max(logits, 1)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x180 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def unnormalize(tensor, mean, std):\n",
    "    for t, m, s in zip(tensor, mean, std):\n",
    "        t.mul_(s).add_(m)\n",
    "    return tensor\n",
    "\n",
    "n_images = 10\n",
    "\n",
    "fig, axes = plt.subplots(nrows=1, ncols=n_images, \n",
    "                         sharex=True, sharey=True, figsize=(20, 2.5))\n",
    "orig_images = features[:n_images]\n",
    "\n",
    "for i in range(n_images):\n",
    "    curr_img = orig_images[i].detach().to(torch.device('cpu'))\n",
    "    curr_img = unnormalize(curr_img,\n",
    "                           torch.tensor([0.485, 0.456, 0.406]),\n",
    "                           torch.tensor([0.229, 0.224, 0.225])) \n",
    "    curr_img = curr_img.permute((1, 2, 0))\n",
    "    axes[i].imshow(curr_img)\n",
    "    axes[i].set_title(classes[predicted_labels[i]])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}