├── .gitignore ├── .gitmodules ├── README.md ├── demo_notebook.ipynb ├── images ├── 2_layer_NN.png ├── tensor.png └── variable.png ├── tutorial_presentation.ipynb └── tutorial_presentation.slides.html /.gitignore: -------------------------------------------------------------------------------- 1 | .ipynb_checkpoints 2 | pretrained/logreg.t7 3 | pretrained/MNIST_net.t7 4 | data/raw 5 | data/processed 6 | -------------------------------------------------------------------------------- /.gitmodules: -------------------------------------------------------------------------------- 1 | [submodule "pretrained"] 2 | path = pretrained 3 | url = git@github.com:michael-isaev/surface_models.git 4 | [submodule "data/surface_quality"] 5 | path = data/surface_quality 6 | url = git@github.com:michael-isaev/surface_dataset.git 7 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # accelerated_dl_pytorch 2 | This is the repository for **Accelerated Deep Learning with Pytotch** tutorial and [**Jupyter Day Atlanta 2018**](https://github.com/atl-jugheads/jupyter-day-atlanta-ii) talk slides. It features full tutorial notebook, Jupyter Notebook Slides html file, and a demo with surface finish quality inspection. 3 | 4 | ## Knowledge Prerequisites 5 | This tutorial assumes familiarity with Python and Numpy. 6 | 7 | ## Tutorial Prerequisites 8 | Python3 is required to run this tutorial. You also will need some libraries from SciPy package (NumPy, Matplotlib, Pandas), Jupyter Notebook support, Seaborn for plotting, and Pytorch 0.3.0 or newer. 9 | 10 | The simpliest way to maintain Python with all these libraries as well as many others is to install [Anaconda](https://www.anaconda.com/download). You can Find Pytorch installation instructions on the [Pytorch page](http://pytorch.org). 11 | 12 | CUDA availability is not strictly required, but highly desirable. Life is short -- use a GPU! 13 | 14 | ## How to Use 15 | Our tutorial git has two submodules - for surface dataset, and for pretrained model for surface finish quality inspection. To download the tutorial, use 16 | 17 | ``` 18 | git clone --recurse-submodules git@github.com:hpcgarage/accelerated_dl_pytorch.git 19 | ``` 20 | 21 | If you didn't clone repository with its submodules, you can always clone submodules with this command: 22 | 23 | ``` 24 | git submodule update --init --recursive 25 | ``` 26 | 27 | To run the Jupyter Notebook Slides as at Jupyter Day Atlanta 2018 talk, you can use following command: 28 | 29 | ``` 30 | jupyter nbconvert tutorial_presentation.ipynb --to slides --post serve 31 | ``` 32 | 33 | ## Table of Contents 34 | 1. Essential PyTorch Background 35 | 2. PyTorch for Data Analytics 36 | 3. LeNet Convolutional Neural Network (CNN) in PyTorch 37 | 4. Application to a Manufacturing Problem 38 | 39 | -------------------------------------------------------------------------------- /demo_notebook.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 15, 6 | "metadata": { 7 | "collapsed": true 8 | }, 9 | "outputs": [], 10 | "source": [ 11 | "import numpy as np\n", 12 | "import torch\n", 13 | "from torch.autograd import Variable\n", 14 | "import torch.nn.functional as F\n", 15 | "import torch.nn as nn\n", 16 | "import torch.optim as optim\n", 17 | "from torch.optim import lr_scheduler\n", 18 | "from skimage import io, transform\n", 19 | "from matplotlib import pyplot as plt\n", 20 | "import torchvision\n", 21 | "from torchvision import models\n", 22 | "from torch.utils.data import Dataset, DataLoader\n", 23 | "from torchvision import datasets, transforms\n", 24 | "import os\n", 25 | "plt.ion() # interactive mode" 26 | ] 27 | }, 28 | { 29 | "cell_type": "code", 30 | "execution_count": 16, 31 | "metadata": {}, 32 | "outputs": [ 33 | { 34 | "name": "stdout", 35 | "output_type": "stream", 36 | "text": [ 37 | "True\n" 38 | ] 39 | } 40 | ], 41 | "source": [ 42 | "print(torch.cuda.is_available())" 43 | ] 44 | }, 45 | { 46 | "cell_type": "code", 47 | "execution_count": 17, 48 | "metadata": { 49 | "collapsed": true 50 | }, 51 | "outputs": [], 52 | "source": [ 53 | "# Load the test dataset and normalize it\n", 54 | "\n", 55 | "data_transforms = {\n", 56 | " 'test': transforms.Compose([\n", 57 | " transforms.Resize(256),\n", 58 | " transforms.CenterCrop(224),\n", 59 | " transforms.ToTensor(),\n", 60 | " transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n", 61 | " ]),\n", 62 | "}\n", 63 | "\n", 64 | "data_dir = 'data/surface_quality'\n", 65 | "test_dataset = datasets.ImageFolder(os.path.join(data_dir, 'test'),\n", 66 | " data_transforms['test'])\n", 67 | "dataloader = torch.utils.data.DataLoader(test_dataset, batch_size=64,\n", 68 | " shuffle=True, num_workers=4)\n", 69 | "dataset_size = len(test_dataset)\n", 70 | "class_names = test_dataset.classes" 71 | ] 72 | }, 73 | { 74 | "cell_type": "code", 75 | "execution_count": 18, 76 | "metadata": { 77 | "collapsed": true 78 | }, 79 | "outputs": [], 80 | "source": [ 81 | "def test(model, epoch, criterion, data_loader):\n", 82 | " model.eval()\n", 83 | " test_loss = 0\n", 84 | " correct = 0\n", 85 | " for data, target in data_loader:\n", 86 | " if torch.cuda.is_available():\n", 87 | " data, target = data.cuda(), target.cuda()\n", 88 | " model.cuda()\n", 89 | " data, target = Variable(data), Variable(target)\n", 90 | " output = model(data)\n", 91 | " test_loss += criterion(output, target).data[0]\n", 92 | " pred = output.data.max(1)[1] # get the index of the max log-probability\n", 93 | " correct += pred.eq(target.data).cpu().sum()\n", 94 | "\n", 95 | " test_loss /= len(data_loader) # loss function already averages over batch size\n", 96 | " acc = correct / len(data_loader.dataset)\n", 97 | " print('\\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\\n'.format(\n", 98 | " test_loss, correct, len(data_loader.dataset), 100. * acc))\n", 99 | " return (acc, test_loss)" 100 | ] 101 | }, 102 | { 103 | "cell_type": "code", 104 | "execution_count": 19, 105 | "metadata": { 106 | "collapsed": true 107 | }, 108 | "outputs": [], 109 | "source": [ 110 | "def imshow(batch, class_names=None, num_images=4):\n", 111 | " plt.figure(figsize=(1.7*num_images, 1.7))\n", 112 | " img, classes = batch\n", 113 | " img_num = min(num_images, img.shape[0])\n", 114 | "\n", 115 | " grid = torchvision.utils.make_grid(img[:img_num], nrow=img_num, padding=10, pad_value=10)\n", 116 | " grid = grid.cpu().numpy().transpose((1, 2, 0))\n", 117 | " mean = np.array([0.485, 0.456, 0.406])\n", 118 | " std = np.array([0.229, 0.224, 0.225])\n", 119 | " grid = std * grid + mean\n", 120 | " grid = np.clip(grid, 0, 1)\n", 121 | " plt.imshow(grid)\n", 122 | " if class_names:\n", 123 | " titles = [class_names[x] for x in classes[:img_num]]\n", 124 | " plt.title(titles)\n", 125 | " plt.pause(0.001) # pause a bit so that plots are updated" 126 | ] 127 | }, 128 | { 129 | "cell_type": "code", 130 | "execution_count": 20, 131 | "metadata": { 132 | "collapsed": true 133 | }, 134 | "outputs": [], 135 | "source": [ 136 | "def visualize_model(model, dataset, class_names=None, num_images=6):\n", 137 | " inputs, target = next(iter(dataset))\n", 138 | " inputs, target = Variable(inputs), Variable(target)\n", 139 | " if torch.cuda.is_available():\n", 140 | " inputs, target = inputs.cuda(), target.cuda()\n", 141 | " model.cuda()\n", 142 | " outputs = model(inputs)\n", 143 | " _, preds = torch.max(outputs.data, 1)\n", 144 | " imshow((inputs.data, preds), class_names=class_names, num_images=num_images)\n", 145 | " if class_names:\n", 146 | " print ('Actual labels:\\n',[class_names[x] for x in target.data[:num_images]])\n", 147 | " " 148 | ] 149 | }, 150 | { 151 | "cell_type": "code", 152 | "execution_count": 21, 153 | "metadata": { 154 | "collapsed": true 155 | }, 156 | "outputs": [], 157 | "source": [ 158 | "model = torch.load('pretrained/resnet18_surface.t7', map_location=lambda storage, loc: storage)" 159 | ] 160 | }, 161 | { 162 | "cell_type": "code", 163 | "execution_count": 12, 164 | "metadata": {}, 165 | "outputs": [ 166 | { 167 | "name": "stdout", 168 | "output_type": "stream", 169 | "text": [ 170 | "ResNet(\n", 171 | " (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)\n", 172 | " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True)\n", 173 | " (relu): ReLU(inplace)\n", 174 | " (maxpool): MaxPool2d(kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), ceil_mode=False)\n", 175 | " (layer1): Sequential(\n", 176 | " (0): BasicBlock(\n", 177 | " (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 178 | " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True)\n", 179 | " (relu): ReLU(inplace)\n", 180 | " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 181 | " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True)\n", 182 | " )\n", 183 | " (1): BasicBlock(\n", 184 | " (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 185 | " (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True)\n", 186 | " (relu): ReLU(inplace)\n", 187 | " (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 188 | " (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True)\n", 189 | " )\n", 190 | " )\n", 191 | " (layer2): Sequential(\n", 192 | " (0): BasicBlock(\n", 193 | " (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", 194 | " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True)\n", 195 | " (relu): ReLU(inplace)\n", 196 | " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 197 | " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True)\n", 198 | " (downsample): Sequential(\n", 199 | " (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", 200 | " (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True)\n", 201 | " )\n", 202 | " )\n", 203 | " (1): BasicBlock(\n", 204 | " (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 205 | " (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True)\n", 206 | " (relu): ReLU(inplace)\n", 207 | " (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 208 | " (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True)\n", 209 | " )\n", 210 | " )\n", 211 | " (layer3): Sequential(\n", 212 | " (0): BasicBlock(\n", 213 | " (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", 214 | " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True)\n", 215 | " (relu): ReLU(inplace)\n", 216 | " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 217 | " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True)\n", 218 | " (downsample): Sequential(\n", 219 | " (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", 220 | " (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True)\n", 221 | " )\n", 222 | " )\n", 223 | " (1): BasicBlock(\n", 224 | " (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 225 | " (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True)\n", 226 | " (relu): ReLU(inplace)\n", 227 | " (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 228 | " (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True)\n", 229 | " )\n", 230 | " )\n", 231 | " (layer4): Sequential(\n", 232 | " (0): BasicBlock(\n", 233 | " (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n", 234 | " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True)\n", 235 | " (relu): ReLU(inplace)\n", 236 | " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 237 | " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True)\n", 238 | " (downsample): Sequential(\n", 239 | " (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)\n", 240 | " (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True)\n", 241 | " )\n", 242 | " )\n", 243 | " (1): BasicBlock(\n", 244 | " (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 245 | " (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True)\n", 246 | " (relu): ReLU(inplace)\n", 247 | " (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n", 248 | " (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True)\n", 249 | " )\n", 250 | " )\n", 251 | " (avgpool): AvgPool2d(kernel_size=7, stride=1, padding=0, ceil_mode=False, count_include_pad=True)\n", 252 | " (fc): Linear(in_features=512, out_features=2, bias=True)\n", 253 | ")\n" 254 | ] 255 | } 256 | ], 257 | "source": [ 258 | "print(model)" 259 | ] 260 | }, 261 | { 262 | "cell_type": "code", 263 | "execution_count": 22, 264 | "metadata": {}, 265 | "outputs": [ 266 | { 267 | "name": "stdout", 268 | "output_type": "stream", 269 | "text": [ 270 | "\n", 271 | "Test set: Average loss: 0.0743, Accuracy: 136/140 (97%)\n", 272 | "\n" 273 | ] 274 | } 275 | ], 276 | "source": [ 277 | "#Testing the cnn\n", 278 | "criterion = nn.CrossEntropyLoss()\n", 279 | "acc, loss = test(model, 0, criterion, dataloader)" 280 | ] 281 | }, 282 | { 283 | "cell_type": "code", 284 | "execution_count": 23, 285 | "metadata": {}, 286 | "outputs": [ 287 | { 288 | "data": { 289 | "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkkAAACLCAYAAAB1Pw2iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n", 290 | "text/plain": [ 291 | "" 292 | ] 293 | }, 294 | "metadata": {}, 295 | "output_type": "display_data" 296 | }, 297 | { 298 | "name": "stdout", 299 | "output_type": "stream", 300 | "text": [ 301 | "Actual labels:\n", 302 | " ['good_surface', 'bad_surface', 'bad_surface', 'good_surface', 'good_surface', 'good_surface']\n" 303 | ] 304 | } 305 | ], 306 | "source": [ 307 | "visualize_model(model, dataloader, num_images=6, class_names=class_names)" 308 | ] 309 | }, 310 | { 311 | "cell_type": "code", 312 | "execution_count": null, 313 | "metadata": { 314 | "collapsed": true 315 | }, 316 | "outputs": [], 317 | "source": [] 318 | } 319 | ], 320 | "metadata": { 321 | "kernelspec": { 322 | "display_name": "Python 3", 323 | "language": "python", 324 | "name": "python3" 325 | }, 326 | "language_info": { 327 | "codemirror_mode": { 328 | "name": "ipython", 329 | "version": 3 330 | }, 331 | "file_extension": ".py", 332 | "mimetype": "text/x-python", 333 | "name": "python", 334 | "nbconvert_exporter": "python", 335 | "pygments_lexer": "ipython3", 336 | "version": "3.6.3" 337 | } 338 | }, 339 | "nbformat": 4, 340 | "nbformat_minor": 2 341 | } 342 | -------------------------------------------------------------------------------- /images/2_layer_NN.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hpcgarage/accelerated_dl_pytorch/8e4e3af712fb1df510388f9d9324a3c59f6c1401/images/2_layer_NN.png -------------------------------------------------------------------------------- /images/tensor.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hpcgarage/accelerated_dl_pytorch/8e4e3af712fb1df510388f9d9324a3c59f6c1401/images/tensor.png -------------------------------------------------------------------------------- /images/variable.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/hpcgarage/accelerated_dl_pytorch/8e4e3af712fb1df510388f9d9324a3c59f6c1401/images/variable.png --------------------------------------------------------------------------------