classifying-cifar-10-images-with-densenet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_kg_hide-input": true,
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [],
   "source": [
    "# Imports\n",
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "\n",
    "# import utilities\n",
    "import os\n",
    "import time\n",
    "\n",
    "# import data visualization\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# import pytorch\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from torch.optim import SGD,Adam,lr_scheduler\n",
    "from torch.utils.data import random_split\n",
    "import torchvision\n",
    "from torchvision import transforms, datasets\n",
    "from torch.utils.data import DataLoader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [],
   "source": [
    "# define transformations for train\n",
    "train_transform = transforms.Compose([\n",
    "    transforms.RandomHorizontalFlip(p=.40),\n",
    "    transforms.RandomRotation(30),\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])])\n",
    "\n",
    "# define transformations for test\n",
    "test_transform = transforms.Compose([\n",
    "    transforms.ToTensor(),\n",
    "    transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])])\n",
    "\n",
    "# define training dataloader\n",
    "def get_training_dataloader(train_transform, batch_size=128, num_workers=0, shuffle=True):\n",
    "    \"\"\" return training dataloader\n",
    "    Args:\n",
    "        train_transform: transfroms for train dataset\n",
    "        path: path to cifar100 training python dataset\n",
    "        batch_size: dataloader batchsize\n",
    "        num_workers: dataloader num_works\n",
    "        shuffle: whether to shuffle \n",
    "    Returns: train_data_loader:torch dataloader object\n",
    "    \"\"\"\n",
    "\n",
    "    transform_train = train_transform\n",
    "    cifar10_training = torchvision.datasets.CIFAR10(root='.', train=True, download=True, transform=transform_train)\n",
    "    cifar10_training_loader = DataLoader(\n",
    "        cifar10_training, shuffle=shuffle, num_workers=num_workers, batch_size=batch_size)\n",
    "\n",
    "    return cifar10_training_loader\n",
    "\n",
    "# define test dataloader\n",
    "def get_testing_dataloader(test_transform, batch_size=128, num_workers=0, shuffle=True):\n",
    "    \"\"\" return training dataloader\n",
    "    Args:\n",
    "        test_transform: transforms for test dataset\n",
    "        path: path to cifar100 test python dataset\n",
    "        batch_size: dataloader batchsize\n",
    "        num_workers: dataloader num_works\n",
    "        shuffle: whether to shuffle \n",
    "    Returns: cifar100_test_loader:torch dataloader object\n",
    "    \"\"\"\n",
    "\n",
    "    transform_test = test_transform\n",
    "    cifar10_test = torchvision.datasets.CIFAR10(root='.', train=False, download=True, transform=transform_test)\n",
    "    cifar10_test_loader = DataLoader(cifar10_test, shuffle=shuffle, num_workers=num_workers, batch_size=batch_size)\n",
    "\n",
    "    return cifar10_test_loader"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "0it [00:00, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./cifar-10-python.tar.gz\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "170500096it [00:08, 20616733.54it/s]                               \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "# get dataloaders for training and testing datasets\n",
    "# don't forget to turn on internet in kernel's settings\n",
    "trainloader = get_training_dataloader(train_transform)\n",
    "testloader = get_testing_dataloader(test_transform)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# specify the names of the classes\n",
    "classes_dict = {0 : 'airplane', 1 : 'automobile', 2: 'bird', 3 : 'cat', 4 : 'deer', 5: 'dog', 6:'frog', 7 : 'horse', 8 : 'ship', 9 : 'truck'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 25 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot 25 random images from training dataset\n",
    "fig, axs = plt.subplots(5, 5, figsize=(10,10))\n",
    "    \n",
    "for batch_idx, (inputs, labels) in enumerate(trainloader):\n",
    "    for im in range(25):\n",
    "        image = inputs[im].permute(1, 2, 0)\n",
    "        i = im // 5\n",
    "        j = im % 5\n",
    "        axs[i,j].imshow(image.numpy()) #plot the data\n",
    "        axs[i,j].axis('off')\n",
    "        axs[i,j].set_title(classes_dict[int(labels[im].numpy())])\n",
    "        \n",
    "    break;\n",
    "\n",
    "# set suptitle\n",
    "plt.suptitle('CIFAR-10 Images')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "outputs": [],
   "source": [
    "#\"\"\"Bottleneck layers. Although each layer only produces k\n",
    "#output feature-maps, it typically has many more inputs. It\n",
    "#has been noted in [37, 11] that a 1×1 convolution can be in-\n",
    "#troduced as bottleneck layer before each 3×3 convolution\n",
    "#to reduce the number of input feature-maps, and thus to\n",
    "#improve computational efficiency.\"\"\"\n",
    "class Bottleneck(nn.Module):\n",
    "    def __init__(self, in_channels, growth_rate):\n",
    "        super().__init__()\n",
    "        #\"\"\"In  our experiments, we let each 1×1 convolution \n",
    "        #produce 4k feature-maps.\"\"\"\n",
    "        inner_channel = 4 * growth_rate\n",
    "\n",
    "        #\"\"\"We find this design especially effective for DenseNet and \n",
    "        #we refer to our network with such a bottleneck layer, i.e., \n",
    "        #to the BN-ReLU-Conv(1×1)-BN-ReLU-Conv(3×3) version of H ` , \n",
    "        #as DenseNet-B.\"\"\"\n",
    "        self.bottle_neck = nn.Sequential(\n",
    "            nn.BatchNorm2d(in_channels),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(in_channels, inner_channel, kernel_size=1, bias=False),\n",
    "            nn.BatchNorm2d(inner_channel),\n",
    "            nn.ReLU(inplace=True),\n",
    "            nn.Conv2d(inner_channel, growth_rate, kernel_size=3, padding=1, bias=False)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        return torch.cat([x, self.bottle_neck(x)], 1)\n",
    "\n",
    "#\"\"\"We refer to layers between blocks as transition\n",
    "#layers, which do convolution and pooling.\"\"\"\n",
    "class Transition(nn.Module):\n",
    "    def __init__(self, in_channels, out_channels):\n",
    "        super().__init__()\n",
    "        #\"\"\"The transition layers used in our experiments \n",
    "        #consist of a batch normalization layer and an 1×1 \n",
    "        #convolutional layer followed by a 2×2 average pooling \n",
    "        #layer\"\"\".\n",
    "        self.down_sample = nn.Sequential(\n",
    "            nn.BatchNorm2d(in_channels),\n",
    "            nn.Conv2d(in_channels, out_channels, 1, bias=False),\n",
    "            nn.AvgPool2d(2, stride=2)\n",
    "        )\n",
    "\n",
    "    def forward(self, x):\n",
    "        return self.down_sample(x)\n",
    "\n",
    "#DesneNet-BC\n",
    "#B stands for bottleneck layer(BN-RELU-CONV(1x1)-BN-RELU-CONV(3x3))\n",
    "#C stands for compression factor(0<=theta<=1)\n",
    "class DenseNet(nn.Module):\n",
    "    def __init__(self, block, nblocks, growth_rate=12, reduction=0.5, num_class=10):\n",
    "        super().__init__()\n",
    "        self.growth_rate = growth_rate\n",
    "\n",
    "        #\"\"\"Before entering the first dense block, a convolution \n",
    "        #with 16 (or twice the growth rate for DenseNet-BC) \n",
    "        #output channels is performed on the input images.\"\"\"\n",
    "        inner_channels = 2 * growth_rate\n",
    "\n",
    "        #For convolutional layers with kernel size 3×3, each \n",
    "        #side of the inputs is zero-padded by one pixel to keep \n",
    "        #the feature-map size fixed.\n",
    "        self.conv1 = nn.Conv2d(3, inner_channels, kernel_size=3, padding=1, bias=False) \n",
    "\n",
    "        self.features = nn.Sequential()\n",
    "\n",
    "        for index in range(len(nblocks) - 1):\n",
    "            self.features.add_module(\"dense_block_layer_{}\".format(index), self._make_dense_layers(block, inner_channels, nblocks[index]))\n",
    "            inner_channels += growth_rate * nblocks[index]\n",
    "\n",
    "            #\"\"\"If a dense block contains m feature-maps, we let the \n",
    "            #following transition layer generate θm output feature-\n",
    "            #maps, where 0 < θ ≤ 1 is referred to as the compression \n",
    "            #fac-tor.\n",
    "            out_channels = int(reduction * inner_channels) # int() will automatic floor the value\n",
    "            self.features.add_module(\"transition_layer_{}\".format(index), Transition(inner_channels, out_channels))\n",
    "            inner_channels = out_channels\n",
    "\n",
    "        self.features.add_module(\"dense_block{}\".format(len(nblocks) - 1), self._make_dense_layers(block, inner_channels, nblocks[len(nblocks)-1]))\n",
    "        inner_channels += growth_rate * nblocks[len(nblocks) - 1]\n",
    "        self.features.add_module('bn', nn.BatchNorm2d(inner_channels))\n",
    "        self.features.add_module('activation', nn.ReLU(inplace=True))\n",
    "\n",
    "        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))\n",
    "\n",
    "        self.linear = nn.Linear(inner_channels, num_class)\n",
    "\n",
    "    def forward(self, x):\n",
    "        output = self.conv1(x)\n",
    "        output = self.features(output)\n",
    "        output = self.avgpool(output)\n",
    "        output = output.view(output.size()[0], -1)\n",
    "        output = self.linear(output)\n",
    "        return output\n",
    "\n",
    "    def _make_dense_layers(self, block, in_channels, nblocks):\n",
    "        dense_block = nn.Sequential()\n",
    "        for index in range(nblocks):\n",
    "            dense_block.add_module('bottle_neck_layer_{}'.format(index), block(in_channels, self.growth_rate))\n",
    "            in_channels += self.growth_rate\n",
    "        return dense_block\n",
    "\n",
    "def densenet121(activation = 'relu'):\n",
    "    return DenseNet(Bottleneck, [6,12,24,16], growth_rate=32)\n",
    "\n",
    "def densenet169(activation = 'relu'):\n",
    "    return DenseNet(Bottleneck, [6,12,32,32], growth_rate=32)\n",
    "\n",
    "def densenet201(activation = 'relu'):\n",
    "    return DenseNet(Bottleneck, [6,12,48,32], growth_rate=32)\n",
    "\n",
    "def densenet161(activation = 'relu'):\n",
    "    return DenseNet(Bottleneck, [6,12,36,24], growth_rate=48)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='cuda', index=0)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number of epochs\n",
    "epochs = 50\n",
    "# learning rate\n",
    "learning_rate = 0.001\n",
    "# device to use\n",
    "# don't forget to turn on GPU on kernel's settings\n",
    "device = torch.device('cuda:0' if torch.cuda.is_available() else \"cpu\")\n",
    "device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = densenet121()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set loss function\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "\n",
    "# set optimizer, only train the classifier parameters, feature parameters are frozen\n",
    "optimizer = Adam(model.parameters(), lr=learning_rate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_stats = pd.DataFrame(columns = ['Epoch', 'Time per epoch', 'Avg time per step', 'Train loss', 'Train accuracy', 'Train top-3 accuracy','Test loss', 'Test accuracy', 'Test top-3 accuracy']) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_kg_hide-output": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50.. Time per epoch: 112.2941.. Average time per step: 0.2872.. Train loss: 1.4506.. Train accuracy: 0.4649.. Top-3 train accuracy: 0.7967.. Test loss: 1.2923.. Test accuracy: 0.5354.. Top-3 test accuracy: 0.8430\n",
      "Epoch 2/50.. Time per epoch: 111.8954.. Average time per step: 0.2862.. Train loss: 1.0713.. Train accuracy: 0.6136.. Top-3 train accuracy: 0.8874.. Test loss: 1.1608.. Test accuracy: 0.5989.. Top-3 test accuracy: 0.8824\n",
      "Epoch 3/50.. Time per epoch: 111.8507.. Average time per step: 0.2861.. Train loss: 0.9035.. Train accuracy: 0.6780.. Top-3 train accuracy: 0.9151.. Test loss: 0.8575.. Test accuracy: 0.6960.. Top-3 test accuracy: 0.9203\n",
      "Epoch 4/50.. Time per epoch: 111.8734.. Average time per step: 0.2861.. Train loss: 0.7815.. Train accuracy: 0.7238.. Top-3 train accuracy: 0.9342.. Test loss: 0.7058.. Test accuracy: 0.7578.. Top-3 test accuracy: 0.9442\n",
      "Epoch 5/50.. Time per epoch: 111.7798.. Average time per step: 0.2859.. Train loss: 0.6943.. Train accuracy: 0.7572.. Top-3 train accuracy: 0.9454.. Test loss: 0.6002.. Test accuracy: 0.7896.. Top-3 test accuracy: 0.9577\n",
      "Epoch 6/50.. Time per epoch: 111.8041.. Average time per step: 0.2859.. Train loss: 0.6213.. Train accuracy: 0.7843.. Top-3 train accuracy: 0.9533.. Test loss: 0.5865.. Test accuracy: 0.7979.. Top-3 test accuracy: 0.9574\n",
      "Epoch 7/50.. Time per epoch: 111.8723.. Average time per step: 0.2861.. Train loss: 0.5608.. Train accuracy: 0.8036.. Top-3 train accuracy: 0.9617.. Test loss: 0.5343.. Test accuracy: 0.8179.. Top-3 test accuracy: 0.9654\n",
      "Epoch 8/50.. Time per epoch: 111.9952.. Average time per step: 0.2864.. Train loss: 0.5157.. Train accuracy: 0.8205.. Top-3 train accuracy: 0.9664.. Test loss: 0.5343.. Test accuracy: 0.8097.. Top-3 test accuracy: 0.9691\n",
      "Epoch 9/50.. Time per epoch: 111.8416.. Average time per step: 0.2860.. Train loss: 0.4804.. Train accuracy: 0.8320.. Top-3 train accuracy: 0.9700.. Test loss: 0.4646.. Test accuracy: 0.8421.. Top-3 test accuracy: 0.9722\n",
      "Epoch 10/50.. Time per epoch: 111.8655.. Average time per step: 0.2861.. Train loss: 0.4416.. Train accuracy: 0.8462.. Top-3 train accuracy: 0.9740.. Test loss: 0.4671.. Test accuracy: 0.8418.. Top-3 test accuracy: 0.9719\n",
      "Epoch 11/50.. Time per epoch: 111.8540.. Average time per step: 0.2861.. Train loss: 0.4123.. Train accuracy: 0.8553.. Top-3 train accuracy: 0.9769.. Test loss: 0.4510.. Test accuracy: 0.8465.. Top-3 test accuracy: 0.9742\n",
      "Epoch 12/50.. Time per epoch: 111.5631.. Average time per step: 0.2853.. Train loss: 0.3882.. Train accuracy: 0.8628.. Top-3 train accuracy: 0.9788.. Test loss: 0.4013.. Test accuracy: 0.8624.. Top-3 test accuracy: 0.9783\n",
      "Epoch 13/50.. Time per epoch: 111.6096.. Average time per step: 0.2854.. Train loss: 0.3574.. Train accuracy: 0.8743.. Top-3 train accuracy: 0.9822.. Test loss: 0.4160.. Test accuracy: 0.8628.. Top-3 test accuracy: 0.9768\n",
      "Epoch 14/50.. Time per epoch: 111.6398.. Average time per step: 0.2855.. Train loss: 0.3333.. Train accuracy: 0.8831.. Top-3 train accuracy: 0.9837.. Test loss: 0.3858.. Test accuracy: 0.8735.. Top-3 test accuracy: 0.9810\n",
      "Epoch 15/50.. Time per epoch: 113.3921.. Average time per step: 0.2900.. Train loss: 0.3190.. Train accuracy: 0.8879.. Top-3 train accuracy: 0.9850.. Test loss: 0.3596.. Test accuracy: 0.8770.. Top-3 test accuracy: 0.9831\n",
      "Epoch 16/50.. Time per epoch: 111.7079.. Average time per step: 0.2857.. Train loss: 0.2959.. Train accuracy: 0.8952.. Top-3 train accuracy: 0.9870.. Test loss: 0.3503.. Test accuracy: 0.8861.. Top-3 test accuracy: 0.9821\n",
      "Epoch 17/50.. Time per epoch: 111.7543.. Average time per step: 0.2858.. Train loss: 0.2786.. Train accuracy: 0.9017.. Top-3 train accuracy: 0.9881.. Test loss: 0.3520.. Test accuracy: 0.8851.. Top-3 test accuracy: 0.9818\n",
      "Epoch 18/50.. Time per epoch: 111.6323.. Average time per step: 0.2855.. Train loss: 0.2615.. Train accuracy: 0.9078.. Top-3 train accuracy: 0.9896.. Test loss: 0.3561.. Test accuracy: 0.8818.. Top-3 test accuracy: 0.9821\n",
      "Epoch 19/50.. Time per epoch: 111.7456.. Average time per step: 0.2858.. Train loss: 0.2453.. Train accuracy: 0.9139.. Top-3 train accuracy: 0.9904.. Test loss: 0.3571.. Test accuracy: 0.8830.. Top-3 test accuracy: 0.9815\n",
      "Epoch 20/50.. Time per epoch: 111.6695.. Average time per step: 0.2856.. Train loss: 0.2295.. Train accuracy: 0.9192.. Top-3 train accuracy: 0.9918.. Test loss: 0.3602.. Test accuracy: 0.8847.. Top-3 test accuracy: 0.9835\n",
      "Epoch 21/50.. Time per epoch: 111.6256.. Average time per step: 0.2855.. Train loss: 0.2217.. Train accuracy: 0.9217.. Top-3 train accuracy: 0.9924.. Test loss: 0.3393.. Test accuracy: 0.8893.. Top-3 test accuracy: 0.9855\n",
      "Epoch 22/50.. Time per epoch: 111.6928.. Average time per step: 0.2857.. Train loss: 0.2063.. Train accuracy: 0.9266.. Top-3 train accuracy: 0.9931.. Test loss: 0.3498.. Test accuracy: 0.8922.. Top-3 test accuracy: 0.9834\n",
      "Epoch 23/50.. Time per epoch: 111.6258.. Average time per step: 0.2855.. Train loss: 0.1939.. Train accuracy: 0.9318.. Top-3 train accuracy: 0.9942.. Test loss: 0.4288.. Test accuracy: 0.8714.. Top-3 test accuracy: 0.9849\n",
      "Epoch 24/50.. Time per epoch: 111.6295.. Average time per step: 0.2855.. Train loss: 0.1850.. Train accuracy: 0.9336.. Top-3 train accuracy: 0.9946.. Test loss: 0.3546.. Test accuracy: 0.8931.. Top-3 test accuracy: 0.9864\n",
      "Epoch 25/50.. Time per epoch: 111.6058.. Average time per step: 0.2854.. Train loss: 0.1727.. Train accuracy: 0.9380.. Top-3 train accuracy: 0.9953.. Test loss: 0.3429.. Test accuracy: 0.8982.. Top-3 test accuracy: 0.9859\n",
      "Epoch 26/50.. Time per epoch: 111.7197.. Average time per step: 0.2857.. Train loss: 0.1669.. Train accuracy: 0.9409.. Top-3 train accuracy: 0.9955.. Test loss: 0.3627.. Test accuracy: 0.8908.. Top-3 test accuracy: 0.9837\n",
      "Epoch 27/50.. Time per epoch: 111.6887.. Average time per step: 0.2856.. Train loss: 0.1571.. Train accuracy: 0.9432.. Top-3 train accuracy: 0.9956.. Test loss: 0.3622.. Test accuracy: 0.8958.. Top-3 test accuracy: 0.9846\n",
      "Epoch 28/50.. Time per epoch: 111.7419.. Average time per step: 0.2858.. Train loss: 0.1476.. Train accuracy: 0.9472.. Top-3 train accuracy: 0.9968.. Test loss: 0.3435.. Test accuracy: 0.8978.. Top-3 test accuracy: 0.9859\n",
      "Epoch 29/50.. Time per epoch: 111.6875.. Average time per step: 0.2856.. Train loss: 0.1371.. Train accuracy: 0.9510.. Top-3 train accuracy: 0.9971.. Test loss: 0.3663.. Test accuracy: 0.8965.. Top-3 test accuracy: 0.9835\n",
      "Epoch 30/50.. Time per epoch: 111.6227.. Average time per step: 0.2855.. Train loss: 0.1353.. Train accuracy: 0.9530.. Top-3 train accuracy: 0.9972.. Test loss: 0.4073.. Test accuracy: 0.8897.. Top-3 test accuracy: 0.9833\n",
      "Epoch 31/50.. Time per epoch: 111.8468.. Average time per step: 0.2861.. Train loss: 0.1264.. Train accuracy: 0.9550.. Top-3 train accuracy: 0.9975.. Test loss: 0.3551.. Test accuracy: 0.8984.. Top-3 test accuracy: 0.9858\n",
      "Epoch 32/50.. Time per epoch: 111.6070.. Average time per step: 0.2854.. Train loss: 0.1185.. Train accuracy: 0.9574.. Top-3 train accuracy: 0.9976.. Test loss: 0.4097.. Test accuracy: 0.8872.. Top-3 test accuracy: 0.9836\n",
      "Epoch 33/50.. Time per epoch: 111.5856.. Average time per step: 0.2854.. Train loss: 0.1152.. Train accuracy: 0.9591.. Top-3 train accuracy: 0.9976.. Test loss: 0.3617.. Test accuracy: 0.8991.. Top-3 test accuracy: 0.9845\n",
      "Epoch 34/50.. Time per epoch: 111.5720.. Average time per step: 0.2854.. Train loss: 0.1100.. Train accuracy: 0.9606.. Top-3 train accuracy: 0.9981.. Test loss: 0.3701.. Test accuracy: 0.8986.. Top-3 test accuracy: 0.9836\n",
      "Epoch 35/50.. Time per epoch: 111.6131.. Average time per step: 0.2855.. Train loss: 0.1006.. Train accuracy: 0.9644.. Top-3 train accuracy: 0.9982.. Test loss: 0.3860.. Test accuracy: 0.8935.. Top-3 test accuracy: 0.9842\n",
      "Epoch 36/50.. Time per epoch: 111.6057.. Average time per step: 0.2854.. Train loss: 0.0969.. Train accuracy: 0.9646.. Top-3 train accuracy: 0.9982.. Test loss: 0.3633.. Test accuracy: 0.9036.. Top-3 test accuracy: 0.9851\n",
      "Epoch 37/50.. Time per epoch: 111.6709.. Average time per step: 0.2856.. Train loss: 0.0987.. Train accuracy: 0.9650.. Top-3 train accuracy: 0.9985.. Test loss: 0.3617.. Test accuracy: 0.9042.. Top-3 test accuracy: 0.9861\n",
      "Epoch 38/50.. Time per epoch: 111.5641.. Average time per step: 0.2853.. Train loss: 0.0947.. Train accuracy: 0.9666.. Top-3 train accuracy: 0.9985.. Test loss: 0.3861.. Test accuracy: 0.9005.. Top-3 test accuracy: 0.9857\n",
      "Epoch 39/50.. Time per epoch: 111.7737.. Average time per step: 0.2859.. Train loss: 0.0918.. Train accuracy: 0.9675.. Top-3 train accuracy: 0.9986.. Test loss: 0.4085.. Test accuracy: 0.8981.. Top-3 test accuracy: 0.9863\n",
      "Epoch 40/50.. Time per epoch: 111.5786.. Average time per step: 0.2854.. Train loss: 0.0859.. Train accuracy: 0.9695.. Top-3 train accuracy: 0.9989.. Test loss: 0.3945.. Test accuracy: 0.8993.. Top-3 test accuracy: 0.9850\n",
      "Epoch 41/50.. Time per epoch: 111.8170.. Average time per step: 0.2860.. Train loss: 0.0830.. Train accuracy: 0.9703.. Top-3 train accuracy: 0.9988.. Test loss: 0.3864.. Test accuracy: 0.9037.. Top-3 test accuracy: 0.9837\n",
      "Epoch 42/50.. Time per epoch: 111.6215.. Average time per step: 0.2855.. Train loss: 0.0789.. Train accuracy: 0.9718.. Top-3 train accuracy: 0.9988.. Test loss: 0.3911.. Test accuracy: 0.8989.. Top-3 test accuracy: 0.9854\n",
      "Epoch 43/50.. Time per epoch: 111.6379.. Average time per step: 0.2855.. Train loss: 0.0796.. Train accuracy: 0.9719.. Top-3 train accuracy: 0.9990.. Test loss: 0.4084.. Test accuracy: 0.8978.. Top-3 test accuracy: 0.9847\n",
      "Epoch 44/50.. Time per epoch: 111.6944.. Average time per step: 0.2857.. Train loss: 0.0730.. Train accuracy: 0.9741.. Top-3 train accuracy: 0.9992.. Test loss: 0.3881.. Test accuracy: 0.8998.. Top-3 test accuracy: 0.9849\n",
      "Epoch 45/50.. Time per epoch: 111.7264.. Average time per step: 0.2857.. Train loss: 0.0702.. Train accuracy: 0.9748.. Top-3 train accuracy: 0.9991.. Test loss: 0.4082.. Test accuracy: 0.9006.. Top-3 test accuracy: 0.9863\n",
      "Epoch 46/50.. Time per epoch: 111.6313.. Average time per step: 0.2855.. Train loss: 0.0726.. Train accuracy: 0.9746.. Top-3 train accuracy: 0.9992.. Test loss: 0.3823.. Test accuracy: 0.9054.. Top-3 test accuracy: 0.9868\n",
      "Epoch 47/50.. Time per epoch: 111.7244.. Average time per step: 0.2857.. Train loss: 0.0661.. Train accuracy: 0.9768.. Top-3 train accuracy: 0.9993.. Test loss: 0.3912.. Test accuracy: 0.9023.. Top-3 test accuracy: 0.9876\n",
      "Epoch 48/50.. Time per epoch: 111.5648.. Average time per step: 0.2853.. Train loss: 0.0654.. Train accuracy: 0.9768.. Top-3 train accuracy: 0.9994.. Test loss: 0.3852.. Test accuracy: 0.9075.. Top-3 test accuracy: 0.9871\n",
      "Epoch 49/50.. Time per epoch: 111.6970.. Average time per step: 0.2857.. Train loss: 0.0692.. Train accuracy: 0.9758.. Top-3 train accuracy: 0.9990.. Test loss: 0.3983.. Test accuracy: 0.9066.. Top-3 test accuracy: 0.9856\n",
      "Epoch 50/50.. Time per epoch: 111.6146.. Average time per step: 0.2855.. Train loss: 0.0630.. Train accuracy: 0.9777.. Top-3 train accuracy: 0.9992.. Test loss: 0.4103.. Test accuracy: 0.9058.. Top-3 test accuracy: 0.9866\n"
     ]
    }
   ],
   "source": [
    "#train the model\n",
    "model.to(device)\n",
    "\n",
    "steps = 0\n",
    "running_loss = 0\n",
    "for epoch in range(epochs):\n",
    "    \n",
    "    since = time.time()\n",
    "    \n",
    "    train_accuracy = 0\n",
    "    top3_train_accuracy = 0 \n",
    "    for inputs, labels in trainloader:\n",
    "        steps += 1\n",
    "        # Move input and label tensors to the default device\n",
    "        inputs, labels = inputs.to(device), labels.to(device)\n",
    "        \n",
    "        optimizer.zero_grad()\n",
    "        \n",
    "        logps = model.forward(inputs)\n",
    "        loss = criterion(logps, labels)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        running_loss += loss.item()\n",
    "        \n",
    "        # calculate train top-1 accuracy\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",
    "        train_accuracy += torch.mean(equals.type(torch.FloatTensor)).item()\n",
    "        \n",
    "        # Calculate train top-3 accuracy\n",
    "        np_top3_class = ps.topk(3, dim=1)[1].cpu().numpy()\n",
    "        target_numpy = labels.cpu().numpy()\n",
    "        top3_train_accuracy += np.mean([1 if target_numpy[i] in np_top3_class[i] else 0 for i in range(0, len(target_numpy))])\n",
    "        \n",
    "    time_elapsed = time.time() - since\n",
    "    \n",
    "    test_loss = 0\n",
    "    test_accuracy = 0\n",
    "    top3_test_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",
    "\n",
    "            test_loss += batch_loss.item()\n",
    "\n",
    "            # Calculate test top-1 accuracy\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",
    "            test_accuracy += torch.mean(equals.type(torch.FloatTensor)).item()\n",
    "            \n",
    "            # Calculate test top-3 accuracy\n",
    "            np_top3_class = ps.topk(3, dim=1)[1].cpu().numpy()\n",
    "            target_numpy = labels.cpu().numpy()\n",
    "            top3_test_accuracy += np.mean([1 if target_numpy[i] in np_top3_class[i] else 0 for i in range(0, len(target_numpy))])\n",
    "\n",
    "    print(f\"Epoch {epoch+1}/{epochs}.. \"\n",
    "          f\"Time per epoch: {time_elapsed:.4f}.. \"\n",
    "          f\"Average time per step: {time_elapsed/len(trainloader):.4f}.. \"\n",
    "          f\"Train loss: {running_loss/len(trainloader):.4f}.. \"\n",
    "          f\"Train accuracy: {train_accuracy/len(trainloader):.4f}.. \"\n",
    "          f\"Top-3 train accuracy: {top3_train_accuracy/len(trainloader):.4f}.. \"\n",
    "          f\"Test loss: {test_loss/len(testloader):.4f}.. \"\n",
    "          f\"Test accuracy: {test_accuracy/len(testloader):.4f}.. \"\n",
    "          f\"Top-3 test accuracy: {top3_test_accuracy/len(testloader):.4f}\")\n",
    "\n",
    "    train_stats = train_stats.append({'Epoch': epoch, 'Time per epoch':time_elapsed, 'Avg time per step': time_elapsed/len(trainloader), 'Train loss' : running_loss/len(trainloader), 'Train accuracy': train_accuracy/len(trainloader), 'Train top-3 accuracy':top3_train_accuracy/len(trainloader),'Test loss' : test_loss/len(testloader), 'Test accuracy': test_accuracy/len(testloader), 'Test top-3 accuracy':top3_test_accuracy/len(testloader)}, ignore_index=True)\n",
    "\n",
    "    running_loss = 0\n",
    "    model.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_stats.to_csv('train_log_DenseNet121.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f195e38e400>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(10,7))\n",
    "ax = plt.axes()\n",
    "\n",
    "plt.title(\"Loss\")\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"Loss\");\n",
    "\n",
    "x = range(1, len(train_stats['Train loss'].values) + 1)\n",
    "ax.plot(x, train_stats['Train loss'].values, '-g', label='train loss');\n",
    "ax.plot(x, train_stats['Test loss'].values, '-b', label='test loss');\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f195e311b38>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(10,7))\n",
    "ax = plt.axes()\n",
    "\n",
    "plt.title(\"Accuracy\")\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"Accuracy\");\n",
    "\n",
    "x = range(1, len(train_stats['Train accuracy'].values) + 1)\n",
    "ax.plot(x, train_stats['Train accuracy'].values, '-g', label='train accuracy');\n",
    "ax.plot(x, train_stats['Test accuracy'].values, '-b', label='test accuracy');\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "_kg_hide-input": true
   },
   "outputs": [],
   "source": [
    "def view_classify(img, ps, title):\n",
    "    \"\"\"\n",
    "    Function for viewing an image and it's predicted classes\n",
    "    with matplotlib.\n",
    "\n",
    "    INPUT:\n",
    "        img - (tensor) image file\n",
    "        ps - (tensor) predicted probabilities for each class\n",
    "        title - (str) string with true label\n",
    "    \"\"\"\n",
    "    ps = ps.data.numpy().squeeze()\n",
    "\n",
    "    fig, (ax1, ax2) = plt.subplots(figsize=(6,9), ncols=2)\n",
    "    image = img.permute(1, 2, 0)\n",
    "    ax1.imshow(image.numpy())\n",
    "    ax1.axis('off')\n",
    "    ax2.barh(np.arange(10), ps)\n",
    "    ax2.set_aspect(0.1)\n",
    "    ax2.set_yticks(np.arange(10))\n",
    "    ax2.set_yticklabels(list(classes_dict.values()), size='small');\n",
    "    ax2.set_title(title)\n",
    "    ax2.set_xlim(0, 1.1)\n",
    "\n",
    "    plt.tight_layout()\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for batch_idx, (inputs, labels) in enumerate(testloader):\n",
    "    inputs, labels = inputs.to(device), labels.to(device)\n",
    "    img = inputs[0]\n",
    "    label_true = labels[0]\n",
    "    ps = model(inputs)\n",
    "    view_classify(img.cpu(), torch.softmax(ps[0].cpu(), dim=0), classes_dict[int(label_true.cpu().numpy())])\n",
    "    \n",
    "    break;"
   ]
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}