Copy of 2237568 Arda Karaman CENG501 - Spring2022 - PA2 - Task 3 - Finetune ResNet.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"Copy of 2237568 Arda Karaman CENG501 - Spring2022 - PA2 - Task 3 - Finetune ResNet.ipynb","provenance":[{"file_id":"1mlvG7QujcGI3kWaIDTSE8mIM3I171R2f","timestamp":1654417213728},{"file_id":"1F6Ja--D3DppGzKCUtaYurxXqOJ_oouqN","timestamp":1652992925796}],"collapsed_sections":[],"toc_visible":true},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU","widgets":{"application/vnd.jupyter.widget-state+json":{"828cf7cdd79847e5ae560e3c75d478de":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_7699131a0110480ba9025857d2f05e17","IPY_MODEL_576f75ddc0154224b85509db48a2f1ac","IPY_MODEL_81c5618855d147f08fc232b58e95b36d"],"layout":"IPY_MODEL_53b10e36cfba4521935e55abf4bbba58"}},"7699131a0110480ba9025857d2f05e17":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_ea6cd81b8aa546c2a4191414aa03aec3","placeholder":"​","style":"IPY_MODEL_9f752464e620487c8ff62a012d971bc6","value":"100%"}},"576f75ddc0154224b85509db48a2f1ac":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_f57ffc245c094ddc9b4351158d757bfc","max":46830571,"min":0,"orientation":"horizontal","style":"IPY_MODEL_35a966d90c1244ccb03e28a090d3b249","value":46830571}},"81c5618855d147f08fc232b58e95b36d":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_e6bd46d5173e4e27a66f7aa31af58f25","placeholder":"​","style":"IPY_MODEL_b93ca1937e2142f3a252bafdf009c895","value":" 44.7M/44.7M [00:00&lt;00:00, 124MB/s]"}},"53b10e36cfba4521935e55abf4bbba58":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"ea6cd81b8aa546c2a4191414aa03aec3":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9f752464e620487c8ff62a012d971bc6":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"f57ffc245c094ddc9b4351158d757bfc":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"35a966d90c1244ccb03e28a090d3b249":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"e6bd46d5173e4e27a66f7aa31af58f25":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"b93ca1937e2142f3a252bafdf009c895":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"0b84244d5c7f44dfb64a1e5ad990ae64":{"model_module":"@jupyter-widgets/controls","model_name":"HBoxModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_989ca34f3ebe4df99a59228ac0535c7a","IPY_MODEL_e19c8713f46b4ba78a6c268509fdd329","IPY_MODEL_daf3abdb0450451f8d32d99c08ed0714"],"layout":"IPY_MODEL_7240cc9d04f74eafa04054ac6fda0587"}},"989ca34f3ebe4df99a59228ac0535c7a":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_7b58134dc7a54a9e940032d906ce3340","placeholder":"​","style":"IPY_MODEL_a4ea9cb6351a4638af3b7e5175cacaf9","value":""}},"e19c8713f46b4ba78a6c268509fdd329":{"model_module":"@jupyter-widgets/controls","model_name":"FloatProgressModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_b3a0d48e3fe146e4978d28d67ddb49a0","max":170498071,"min":0,"orientation":"horizontal","style":"IPY_MODEL_627119ea0dd4489aa848c1851bc393d7","value":170498071}},"daf3abdb0450451f8d32d99c08ed0714":{"model_module":"@jupyter-widgets/controls","model_name":"HTMLModel","model_module_version":"1.5.0","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_2f1b4e85fbbd461a8fc95715af81ca3c","placeholder":"​","style":"IPY_MODEL_819c2dfe94a446a2b418571416bf6487","value":" 170499072/? [00:03&lt;00:00, 69919961.87it/s]"}},"7240cc9d04f74eafa04054ac6fda0587":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7b58134dc7a54a9e940032d906ce3340":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"a4ea9cb6351a4638af3b7e5175cacaf9":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"b3a0d48e3fe146e4978d28d67ddb49a0":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"627119ea0dd4489aa848c1851bc393d7":{"model_module":"@jupyter-widgets/controls","model_name":"ProgressStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"2f1b4e85fbbd461a8fc95715af81ca3c":{"model_module":"@jupyter-widgets/base","model_name":"LayoutModel","model_module_version":"1.2.0","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"819c2dfe94a446a2b418571416bf6487":{"model_module":"@jupyter-widgets/controls","model_name":"DescriptionStyleModel","model_module_version":"1.5.0","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}}}}},"cells":[{"cell_type":"markdown","metadata":{"id":"1X4X2hmaEmip"},"source":["# Task 3: Finetune ResNet\n","# CENG501 - Spring 2022 - PA2\n","\n","In this task, we will practice transfer learning by adapting and finetuning ResNet18 (pretrained on ImageNet) for CIFAR10.\n"]},{"cell_type":"markdown","metadata":{"id":"oRaKrjO5JSu-"},"source":["## 1 Import the Modules\n","\n","Let us start by importing some libraries that we will use throughout the task."]},{"cell_type":"code","metadata":{"id":"v4L5nogMKyNx","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1654416455976,"user_tz":-180,"elapsed":6355,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"04823800-2218-4767-d364-7afb9513e6c1"},"source":["import matplotlib.pyplot as plt # For plotting\n","import numpy as np              # NumPy, for working with arrays/tensors \n","import time                     # For measuring time\n","import random                   # Python's random library\n","\n","# PyTorch libraries:\n","import torch\n","import torch.nn as nn\n","import torchvision\n","import torchvision.transforms as transforms\n","\n","# install and import the torchinfo library\n","!pip install torchinfo\n","from torchinfo import summary\n","\n","%matplotlib inline\n","plt.rcParams['figure.figsize'] = [12, 8]\n","plt.rcParams['figure.dpi'] = 100 # 200 e.g. is really fine, but slower"],"execution_count":1,"outputs":[{"output_type":"stream","name":"stdout","text":["Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n","Collecting torchinfo\n","  Downloading torchinfo-1.7.0-py3-none-any.whl (22 kB)\n","Installing collected packages: torchinfo\n","Successfully installed torchinfo-1.7.0\n"]}]},{"cell_type":"markdown","metadata":{"id":"4BXVQOGRI4Sc"},"source":["### 1.1 Enable GPU\n","\n","From \"Edit -> Notebook Settings -> Hardware accelerator\" select GPU. With the following we will specify to PyTorch that we want to use the GPU."]},{"cell_type":"code","metadata":{"id":"97DJEyArJLcR","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1654416459724,"user_tz":-180,"elapsed":953,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"13ef227a-05be-4e4c-85f1-cbb65a97a788"},"source":["if torch.cuda.is_available():\n","  print(\"Cuda (GPU support) is available and enabled!\")\n","  device = torch.device(\"cuda\")\n","else:\n","  print(\"Cuda (GPU support) is not available :(\")\n","  device = torch.device(\"cpu\")"],"execution_count":2,"outputs":[{"output_type":"stream","name":"stdout","text":["Cuda (GPU support) is available and enabled!\n"]}]},{"cell_type":"markdown","metadata":{"id":"_KyHtEWkOuZW"},"source":["## 2 Download and Test Pretrained ResNet18\n","\n","We have talked about ResNet in detail in the lectures. We will take Pytorch's ResNet18 model (https://pytorch.org/hub/pytorch_vision_resnet/), which was trained on ImageNet, and adapt & finetune it for CIFAR10.\n","\n","Here are a couple of critical information about ResNet18's input (from Pytorch document):\n","\n","\"All pre-trained models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape (3 x H x W), where H and W are expected to be at least 224. The images have to be loaded in to a range of [0, 1] and then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225].\""]},{"cell_type":"markdown","metadata":{"id":"mWxyCVI_OTdi"},"source":["### 2.1 Download ResNet18"]},{"cell_type":"code","metadata":{"id":"iwOkHvOqOuyb","colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["828cf7cdd79847e5ae560e3c75d478de","7699131a0110480ba9025857d2f05e17","576f75ddc0154224b85509db48a2f1ac","81c5618855d147f08fc232b58e95b36d","53b10e36cfba4521935e55abf4bbba58","ea6cd81b8aa546c2a4191414aa03aec3","9f752464e620487c8ff62a012d971bc6","f57ffc245c094ddc9b4351158d757bfc","35a966d90c1244ccb03e28a090d3b249","e6bd46d5173e4e27a66f7aa31af58f25","b93ca1937e2142f3a252bafdf009c895"]},"executionInfo":{"status":"ok","timestamp":1654416464175,"user_tz":-180,"elapsed":733,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"958f4d1c-b163-460f-a129-0c33e70aa976"},"source":["# Thanks to Pytorch datasets, this is very easy:\n","\n","OrigResNet18 = None\n","###########################################################\n","# @TODO: Look at PyTorch's documentation for downloading  #\n","# and loading a pretrained ResNet18 model.                #\n","#                                                         #\n","# Hint: This should be a single function call.            #\n","###########################################################\n","\n","OrigResNet18 = torchvision.models.resnet18(pretrained = True)\n","\n","###########################################################\n","#                         END OF YOUR CODE                #\n","###########################################################\n","print(OrigResNet18)"],"execution_count":3,"outputs":[{"output_type":"stream","name":"stderr","text":["Downloading: \"https://download.pytorch.org/models/resnet18-f37072fd.pth\" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth\n"]},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0.00/44.7M [00:00<?, ?B/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"828cf7cdd79847e5ae560e3c75d478de"}},"metadata":{}},{"output_type":"stream","name":"stdout","text":["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): BasicBlock(\n","      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","    )\n","    (1): BasicBlock(\n","      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","    )\n","  )\n","  (layer2): Sequential(\n","    (0): BasicBlock(\n","      (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","      (downsample): Sequential(\n","        (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)\n","        (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      )\n","    )\n","    (1): BasicBlock(\n","      (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","    )\n","  )\n","  (layer3): Sequential(\n","    (0): BasicBlock(\n","      (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","      (downsample): Sequential(\n","        (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)\n","        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      )\n","    )\n","    (1): BasicBlock(\n","      (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","    )\n","  )\n","  (layer4): Sequential(\n","    (0): BasicBlock(\n","      (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","      (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): BasicBlock(\n","      (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n","      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n","      (relu): ReLU(inplace=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","    )\n","  )\n","  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))\n","  (fc): Linear(in_features=512, out_features=1000, bias=True)\n",")\n"]}]},{"cell_type":"markdown","metadata":{"id":"6bDj9IQRL7ms"},"source":["### 2.2 Get ImageNet Labels\n","\n","Before we finetune ResNet18, let us see what it predicts on CIFAR10. To be able to do so, we will require the list of ImageNet labels."]},{"cell_type":"code","metadata":{"id":"mwRoWkU0L_Cp","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1654416470117,"user_tz":-180,"elapsed":325,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"a4f671c8-4a2f-4677-c444-fa3656a8007b"},"source":["# Download ImageNet labels\n","!wget -nc https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt\n","\n","# Read the categories\n","with open(\"imagenet_classes.txt\", \"r\") as f:\n","    categories = [s.strip() for s in f.readlines()]"],"execution_count":4,"outputs":[{"output_type":"stream","name":"stdout","text":["--2022-06-05 08:07:49--  https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt\n","Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n","Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n","HTTP request sent, awaiting response... 200 OK\n","Length: 10472 (10K) [text/plain]\n","Saving to: ‘imagenet_classes.txt’\n","\n","\rimagenet_classes.tx   0%[                    ]       0  --.-KB/s               \rimagenet_classes.tx 100%[===================>]  10.23K  --.-KB/s    in 0s      \n","\n","2022-06-05 08:07:50 (88.7 MB/s) - ‘imagenet_classes.txt’ saved [10472/10472]\n","\n"]}]},{"cell_type":"markdown","metadata":{"id":"fk6K0jC_OZs0"},"source":["### 2.3 Load CIFAR10\n","\n","As we mentioned above, ResNet18 expects images with resolution 224x224 and normalized with a mean & std. Therefore, while loading CIFAR10, we will apply certain transformations to handle these requirements."]},{"cell_type":"code","metadata":{"id":"7Udg6NjvPU4a","colab":{"base_uri":"https://localhost:8080/","height":100,"referenced_widgets":["0b84244d5c7f44dfb64a1e5ad990ae64","989ca34f3ebe4df99a59228ac0535c7a","e19c8713f46b4ba78a6c268509fdd329","daf3abdb0450451f8d32d99c08ed0714","7240cc9d04f74eafa04054ac6fda0587","7b58134dc7a54a9e940032d906ce3340","a4ea9cb6351a4638af3b7e5175cacaf9","b3a0d48e3fe146e4978d28d67ddb49a0","627119ea0dd4489aa848c1851bc393d7","2f1b4e85fbbd461a8fc95715af81ca3c","819c2dfe94a446a2b418571416bf6487"]},"executionInfo":{"status":"ok","timestamp":1654416480445,"user_tz":-180,"elapsed":7891,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"478bcf5d-60b2-4174-b26e-25a289a44a06"},"source":["batch_size = 8\n","\n","TF = transforms.Compose([\n","    transforms.Resize(256),\n","    transforms.CenterCrop(224),\n","    transforms.ToTensor(),\n","    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n","])\n","\n","trainset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=TF)\n","trainloader = torch.utils.data.DataLoader(trainset, batch_size=batch_size,\n","                                          shuffle=True, num_workers=2)\n","\n","testset = torchvision.datasets.CIFAR10(root='./data', train=False, download=True, transform=TF)\n","testloader = torch.utils.data.DataLoader(testset, batch_size=batch_size,\n","                                         shuffle=False, num_workers=2)\n","\n","CIFAR10_classes = ('plane', 'car', 'bird', 'cat',\n","           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')"],"execution_count":5,"outputs":[{"output_type":"stream","name":"stdout","text":["Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz\n"]},{"output_type":"display_data","data":{"text/plain":["  0%|          | 0/170498071 [00:00<?, ?it/s]"],"application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"0b84244d5c7f44dfb64a1e5ad990ae64"}},"metadata":{}},{"output_type":"stream","name":"stdout","text":["Extracting ./data/cifar-10-python.tar.gz to ./data\n","Files already downloaded and verified\n"]}]},{"cell_type":"markdown","metadata":{"id":"FUxUQKZSO3q2"},"source":["### 2.4 Apply Pre-trained ResNet18 on CIFAR10\n","\n","Let us look at what ResNet18 predicts on a batch of CIFAR10. For a set of samples, we will visualize the images and look at ResNet18's top predictions."]},{"cell_type":"code","metadata":{"id":"A19MMClxQIyr","colab":{"base_uri":"https://localhost:8080/","height":332},"executionInfo":{"status":"ok","timestamp":1654416483587,"user_tz":-180,"elapsed":1470,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"8a75533a-0b05-4eb1-af2b-1a87c8e69192"},"source":["# Get a batch\n","dataiter = iter(testloader)\n","images, labels = next(dataiter)\n","\n","# Get scores for classes on the batch\n","with torch.no_grad():\n","    output = OrigResNet18(images)\n","\n","# Convert them to probabilities (shape: batch_size x 1000)\n","probabilities = torch.nn.functional.softmax(output, dim=1)\n","\n","# Show results on a 2x2 grid\n","S=2\n","for i in range(S):\n","  for j in range(S):\n","    X = images[i*S+j]\n","    X = np.transpose((X.numpy()/2+0.5), (1, 2, 0))\n","    top1_prob, top1_catid = torch.topk(probabilities[i*S+j], 1)\n","    title = \"{} p:{:1.2f}\".format(categories[top1_catid], top1_prob.item())\n","\n","    plt.subplot(S, S, i*S+j+1)\n","    plt.imshow(X)\n","    plt.axis('off')\n","    plt.title(title)\n","    plt.subplots_adjust(hspace = 0.5)\n"],"execution_count":6,"outputs":[{"output_type":"stream","name":"stderr","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"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 4 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"zHmjt73oP3f9"},"source":["We see that the predictions are way off and we will hopefully get better results with some finetuning."]},{"cell_type":"markdown","metadata":{"id":"KV9omiSefxOU"},"source":["## 3 Adapt ResNet18 for CIFAR10\n","\n","We will \"freeze\" the parameters of ResNet18 and replace the last layer of ResNet18 with a new layer which we will finetune."]},{"cell_type":"code","metadata":{"id":"IaL7AMbzZqTw","executionInfo":{"status":"ok","timestamp":1654416487358,"user_tz":-180,"elapsed":262,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}}},"source":["# Copy ResNet18\n","import copy \n","NewResNet18 = copy.deepcopy(OrigResNet18)"],"execution_count":7,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"RTt3F-EHI1_9"},"source":["### 3.1 Freeze Parameters of ResNet18\n","\n","We \"freeze\" a parameter by setting its `requires_grad` member variable to `False`."]},{"cell_type":"code","metadata":{"id":"42z_IRpzINbe","executionInfo":{"status":"ok","timestamp":1654416490357,"user_tz":-180,"elapsed":281,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}}},"source":["###########################################################\n","# @TODO: Go over the parameters of NewResNet18 and set    #\n","# requires_grad for all parameters to False.              #\n","#                                                         #\n","# Hint: Check parameters() member function of NewResNet18.#\n","###########################################################\n","\n","for param in NewResNet18.parameters():\n","    param.requires_grad = False\n","\n","###########################################################\n","#                         END OF YOUR CODE                #\n","###########################################################"],"execution_count":8,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"htbhG_H5I53Z"},"source":["### 3.2 Add a New Learnable FC Layer to ResNet18\n","\n","If you look at the summary of ResNet18 shown above, you will see that the last layer is:\n","\n","  `(fc): Linear(in_features=512, out_features=1000, bias=True)`\n","\n","In our case, we should just replace this with a new FC layer (its dimensions should be straight-forward to figure out)."]},{"cell_type":"code","metadata":{"id":"Eh70CY4gJuiH","executionInfo":{"status":"ok","timestamp":1654416495190,"user_tz":-180,"elapsed":318,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}}},"source":["NewResNet18.fc = None\n","###########################################################\n","# @TODO: Create a new layer and save it into              #\n","# NewResNet18.fc. This new layer will map the activations #\n","# of the previous layer to the outputs for the CIFAR10    #\n","# classes.                                                #\n","###########################################################\n","\n","NewResNet18.fc = nn.Linear(512, 10, bias=True)\n","\n","###########################################################\n","#                         END OF YOUR CODE                #\n","###########################################################"],"execution_count":9,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"eQic000dKCBi"},"source":["### 3.3 Visualize the Model\n","\n","Now, let us see whether the new fc layer is correct for CIFAR10."]},{"cell_type":"code","metadata":{"id":"9opKZmXSKKNV","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1654416498773,"user_tz":-180,"elapsed":303,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"fd27d118-9abe-4731-f226-f7f786761588"},"source":["print(NewResNet18.fc)"],"execution_count":10,"outputs":[{"output_type":"stream","name":"stdout","text":["Linear(in_features=512, out_features=10, bias=True)\n"]}]},{"cell_type":"markdown","metadata":{"id":"E8VqkhyAf1s4"},"source":["## 4 Finetune ResNet18\n","\n","While finetuning ResNet18, we will just update the last layer. "]},{"cell_type":"markdown","metadata":{"id":"h6Xwis1-J26R"},"source":["### 4.1 Training Method\n","\n","This is the same training method from Task 2."]},{"cell_type":"code","metadata":{"id":"GEZU8jZpJ5At","executionInfo":{"status":"ok","timestamp":1654416501231,"user_tz":-180,"elapsed":366,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}}},"source":["def train(model, criterion, optimizer, epochs, dataloader, verbose=True):\n","  \"\"\"\n","    Define the trainer function. We can use this for training any model.\n","    The parameter names are self-explanatory.\n","\n","    Returns: the loss history.\n","  \"\"\"\n","  loss_history = [] \n","  for epoch in range(epochs):\n","    for i, data in enumerate(dataloader, 0):    \n","      \n","      # Our batch:\n","      inputs, labels = data\n","      inputs = inputs.to(device)\n","      labels = labels.to(device)\n","\n","      # zero the gradients as PyTorch accumulates them\n","      optimizer.zero_grad()\n","\n","      # Obtain the scores\n","      outputs = model(inputs)\n","\n","      # Calculate loss\n","      loss = criterion(outputs.to(device), labels)\n","\n","      # Backpropagate\n","      loss.backward()\n","\n","      # Update the weights\n","      optimizer.step()\n","\n","      loss_history.append(loss.item())\n","    \n","    if verbose: print(f'Epoch {epoch} / {epochs}: avg. loss of last 5 iterations {np.sum(loss_history[:-6:-1])/5}')\n","\n","  return loss_history"],"execution_count":11,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"15VCbg0lKPIz"},"source":["### 4.2 Finetune the Adapted ResNet18 on CIFAR10\n","\n","We will only provide the learnable parameters to the optimizer."]},{"cell_type":"code","metadata":{"id":"OfeLlbI4KU_M","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1654416742091,"user_tz":-180,"elapsed":238610,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"7db7a570-7fef-447a-8ecb-53c1de137285"},"source":["import torch.optim as optim\n","\n","# For reproducibility, let us recreate the FC layer here with a fixed seed:\n","torch.manual_seed(501)\n","random.seed(501)\n","np.random.seed(501)    \n","\n","NewResNet18.fc = None\n","###########################################################\n","# @TODO: Repeat what you did in 3.2 here                  #\n","###########################################################\n","\n","NewResNet18.fc = nn.Linear(512, 10, bias=True)\n","\n","###########################################################\n","#                         END OF YOUR CODE                #\n","###########################################################\n","\n","def get_learnable_parameters(model):\n","    params_to_update = []\n","    for name, param in model.named_parameters():\n","        if param.requires_grad == True:\n","            params_to_update.append(param)\n","    return params_to_update\n","\n","parameters_to_update = get_learnable_parameters(NewResNet18)\n","criterion = nn.CrossEntropyLoss()\n","optimizer = optim.SGD(parameters_to_update, lr=0.0001, momentum=0.95)\n","\n","NewResNet18 = NewResNet18.to(device)\n","epochs = 2\n","loss_history = train(NewResNet18, criterion, optimizer, epochs, trainloader)"],"execution_count":12,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 0 / 2: avg. loss of last 5 iterations 0.9939670085906982\n","Epoch 1 / 2: avg. loss of last 5 iterations 0.8708017826080322\n"]}]},{"cell_type":"markdown","metadata":{"id":"3SUT6gj9KenX"},"source":["### 4.3 The Loss Curve\n","\n","You will see that the loss curve is very noisy, which suggests that we should finetune our hyper-parameters. Though, we will see that we get already reasonably well performance on test data."]},{"cell_type":"code","metadata":{"id":"dNzegVoFKgXu","colab":{"base_uri":"https://localhost:8080/","height":279},"executionInfo":{"status":"ok","timestamp":1654416747618,"user_tz":-180,"elapsed":10,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"be6ace79-b8e4-4744-cdf2-b12c787827a6"},"source":["plt.plot(loss_history)\n","plt.xlabel('Iteration number')\n","plt.ylabel('Loss value')\n","plt.show()"],"execution_count":13,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","metadata":{"id":"8cBp7Xu_Kkhp"},"source":["### 4.4 Quantitative Results\n","\n","We can analyze the accuracy of the predictions as follows. You should see around 69\\% accuracies. We can finetune the hyperparameters to obtain better results.\n","\n","*Disclaimer: This code piece is taken from PyTorch examples.*"]},{"cell_type":"code","metadata":{"id":"dz0pvi8AKmse","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1654416773932,"user_tz":-180,"elapsed":22886,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"f7c877b7-e358-4178-f260-101b763abe93"},"source":["correct = 0\n","total = 0\n","\n","NewResNet18 = NewResNet18.to(device)\n","\n","with torch.no_grad():\n","    for data in testloader:\n","        images, labels = data\n","        images = images.to(device)\n","        labels = labels.to(device)\n","        outputs = NewResNet18(images)\n","        _, predicted = torch.max(outputs.data, 1)\n","        total += labels.size(0)\n","        correct += (predicted == labels).sum().item()\n","\n","print('Accuracy of the network on the 10000 test images: %d %%' % (\n","    100 * correct / total))"],"execution_count":14,"outputs":[{"output_type":"stream","name":"stdout","text":["Accuracy of the network on the 10000 test images: 69 %\n"]}]},{"cell_type":"markdown","metadata":{"id":"69QV-5JKVHVl"},"source":["### 4.5 Visual Results\n","\n","We see that with just two epochs of training a single FC layer, we can get decent results."]},{"cell_type":"code","metadata":{"id":"2ALpDTyfVNJd","colab":{"base_uri":"https://localhost:8080/","height":332},"executionInfo":{"status":"ok","timestamp":1654416779177,"user_tz":-180,"elapsed":1151,"user":{"displayName":"Arda Karaman","userId":"09649945184885111594"}},"outputId":"a155ab32-e000-4f22-c501-fffc69d295f4"},"source":["# Get a batch\n","dataiter = iter(testloader)\n","images, labels = next(dataiter)\n","images = images.to(device)\n","labels = labels.to(device)\n","\n","# Get scores for classes on the batch\n","with torch.no_grad():\n","    output = NewResNet18(images)\n","\n","# Convert them to probabilities (shape: batch_size x 1000)\n","probabilities = torch.nn.functional.softmax(output, dim=1)\n","\n","# Show results on a 2x2 grid\n","S=2\n","for i in range(S):\n","  for j in range(S):\n","    X = images[i*S+j]\n","    X = np.transpose((X.to(\"cpu\").numpy()/2+0.5), (1, 2, 0))\n","    top1_prob, top1_catid = torch.topk(probabilities[i*S+j], 1)\n","    title = \"{} p:{:1.2f}\".format(CIFAR10_classes[top1_catid], top1_prob.item())\n","\n","    plt.subplot(S, S, i*S+j+1)\n","    plt.imshow(X)\n","    plt.axis('off')\n","    plt.title(title)\n","    plt.subplots_adjust(hspace = 0.5)\n"],"execution_count":15,"outputs":[{"output_type":"stream","name":"stderr","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"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 4 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]}]}