525 lines
127 KiB
Plaintext
525 lines
127 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"colab_type": "text",
|
|
"id": "view-in-github"
|
|
},
|
|
"source": [
|
|
"<a href=\"https://colab.research.google.com/github/mrdbourke/pytorch-deep-learning/blob/main/extras/exercises/08_pytorch_paper_replicating_exercises.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "zNqPNlYylluR"
|
|
},
|
|
"source": [
|
|
"# 08. PyTorch Paper Replicating Exercises\n",
|
|
"\n",
|
|
"Welcome to the 08. PyTorch Paper Replicating exercises.\n",
|
|
"\n",
|
|
"Your objective is to write code to satisify each of the exercises below.\n",
|
|
"\n",
|
|
"Some starter code has been provided to make sure you have all the resources you need.\n",
|
|
"\n",
|
|
"> **Note:** There may be more than one solution to each of the exercises.\n",
|
|
"\n",
|
|
"## Resources\n",
|
|
"\n",
|
|
"1. These exercises/solutions are based on [section 08. PyTorch Paper Replicating](https://www.learnpytorch.io/08_pytorch_paper_replicating/) of the Learn PyTorch for Deep Learning course by Zero to Mastery.\n",
|
|
"2. See a live [walkthrough of the solutions (errors and all) on YouTube](https://youtu.be/tjpW_BY8y3g) (but try the exercises yourself first!).\n",
|
|
"3. See [all solutions on the course GitHub](https://github.com/mrdbourke/pytorch-deep-learning/tree/main/extras/solutions).\n",
|
|
"\n",
|
|
"> **Note:** The first section of this notebook is dedicated to getting various helper functions and datasets used for the exercises. The exercises start at the heading \"Exercise 1: ...\"."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "sf8ab9cyHTzU"
|
|
},
|
|
"source": [
|
|
"### Get various imports and helper functions\n",
|
|
"\n",
|
|
"The code in the following cells prepares imports and data for the exercises below. They are taken from [08. PyTorch Paper Replicating](https://www.learnpytorch.io/08_pytorch_paper_replicating/). "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/"
|
|
},
|
|
"id": "ChRaHUSJ8DYZ",
|
|
"outputId": "8f1c0a98-c04a-4d4b-a748-e591a315f805"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"torch version: 1.12.0+cu113\n",
|
|
"torchvision version: 0.13.0+cu113\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# For this notebook to run with updated APIs, we need torch 1.12+ and torchvision 0.13+\n",
|
|
"try:\n",
|
|
" import torch\n",
|
|
" import torchvision\n",
|
|
" assert int(torch.__version__.split(\".\")[1]) >= 12, \"torch version should be 1.12+\"\n",
|
|
" assert int(torchvision.__version__.split(\".\")[1]) >= 13, \"torchvision version should be 0.13+\"\n",
|
|
" print(f\"torch version: {torch.__version__}\")\n",
|
|
" print(f\"torchvision version: {torchvision.__version__}\")\n",
|
|
"except:\n",
|
|
" print(f\"[INFO] torch/torchvision versions not as required, installing nightly versions.\")\n",
|
|
" !pip3 install -U --pre torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/nightly/cu113\n",
|
|
" import torch\n",
|
|
" import torchvision\n",
|
|
" print(f\"torch version: {torch.__version__}\")\n",
|
|
" print(f\"torchvision version: {torchvision.__version__}\")\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/"
|
|
},
|
|
"id": "Y5H5P8EjCNGK",
|
|
"outputId": "4214da9e-f3a8-43e6-a48c-1a33f44a9be9"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"[INFO] Couldn't find torchinfo... installing it.\n",
|
|
"[INFO] Couldn't find going_modular or helper_functions scripts... downloading them from GitHub.\n",
|
|
"Cloning into 'pytorch-deep-learning'...\n",
|
|
"remote: Enumerating objects: 2589, done.\u001b[K\n",
|
|
"remote: Counting objects: 100% (28/28), done.\u001b[K\n",
|
|
"remote: Compressing objects: 100% (17/17), done.\u001b[K\n",
|
|
"remote: Total 2589 (delta 11), reused 28 (delta 11), pack-reused 2561\u001b[K\n",
|
|
"Receiving objects: 100% (2589/2589), 446.45 MiB | 41.00 MiB/s, done.\n",
|
|
"Resolving deltas: 100% (1455/1455), done.\n",
|
|
"Checking out files: 100% (184/184), done.\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# Continue with regular imports\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"import torch\n",
|
|
"import torchvision\n",
|
|
"\n",
|
|
"from torch import nn\n",
|
|
"from torchvision import transforms\n",
|
|
"\n",
|
|
"# Try to get torchinfo, install it if it doesn't work\n",
|
|
"try:\n",
|
|
" from torchinfo import summary\n",
|
|
"except:\n",
|
|
" print(\"[INFO] Couldn't find torchinfo... installing it.\")\n",
|
|
" !pip install -q torchinfo\n",
|
|
" from torchinfo import summary\n",
|
|
"\n",
|
|
"# Try to import the going_modular directory, download it from GitHub if it doesn't work\n",
|
|
"try:\n",
|
|
" from going_modular.going_modular import data_setup, engine\n",
|
|
" from helper_functions import download_data, set_seeds, plot_loss_curves\n",
|
|
"except:\n",
|
|
" # Get the going_modular scripts\n",
|
|
" print(\"[INFO] Couldn't find going_modular or helper_functions scripts... downloading them from GitHub.\")\n",
|
|
" !git clone https://github.com/mrdbourke/pytorch-deep-learning\n",
|
|
" !mv pytorch-deep-learning/going_modular .\n",
|
|
" !mv pytorch-deep-learning/helper_functions.py . # get the helper_functions.py script\n",
|
|
" !rm -rf pytorch-deep-learning\n",
|
|
" from going_modular.going_modular import data_setup, engine\n",
|
|
" from helper_functions import download_data, set_seeds, plot_loss_curves"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 35
|
|
},
|
|
"id": "bE1AAH_uCjiP",
|
|
"outputId": "d4f0001d-3854-4845-f195-eeb08c503aa8"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"application/vnd.google.colaboratory.intrinsic+json": {
|
|
"type": "string"
|
|
},
|
|
"text/plain": [
|
|
"'cuda'"
|
|
]
|
|
},
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
|
|
"device"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "GmS5yuvxCpLp"
|
|
},
|
|
"source": [
|
|
"### Get data\n",
|
|
"\n",
|
|
"Want to download the data we've been using in PyTorch Paper Replicating: https://www.learnpytorch.io/08_pytorch_paper_replicating/#1-get-data"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 5,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/"
|
|
},
|
|
"id": "dm772wqgCzN9",
|
|
"outputId": "ca6c646d-dd40-4453-c12c-a1f0ef5f86f4"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"[INFO] Did not find data/pizza_steak_sushi directory, creating one...\n",
|
|
"[INFO] Downloading pizza_steak_sushi.zip from https://github.com/mrdbourke/pytorch-deep-learning/raw/main/data/pizza_steak_sushi.zip...\n",
|
|
"[INFO] Unzipping pizza_steak_sushi.zip data...\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"PosixPath('data/pizza_steak_sushi')"
|
|
]
|
|
},
|
|
"execution_count": 5,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"# Download pizza, steak, sushi images from GitHub\n",
|
|
"image_path = download_data(source=\"https://github.com/mrdbourke/pytorch-deep-learning/raw/main/data/pizza_steak_sushi.zip\",\n",
|
|
" destination=\"pizza_steak_sushi\")\n",
|
|
"image_path"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 6,
|
|
"metadata": {
|
|
"id": "r1ML2c-dCzCi"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# Setup directory paths to train and test images\n",
|
|
"train_dir = image_path / \"train\"\n",
|
|
"test_dir = image_path / \"test\""
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "nNBZ_2h_Cy86"
|
|
},
|
|
"source": [
|
|
"### Preprocess data\n",
|
|
"\n",
|
|
"Turn images into tensors using same code as PyTorch Paper Replicating section 2.1 and 2.2: https://www.learnpytorch.io/08_pytorch_paper_replicating/#21-prepare-transforms-for-images"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 7,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/"
|
|
},
|
|
"id": "mU0T4gP3DJdF",
|
|
"outputId": "de32055d-b807-4245-e4a6-c10653502630"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Manually created transforms: Compose(\n",
|
|
" Resize(size=(224, 224), interpolation=bilinear, max_size=None, antialias=None)\n",
|
|
" ToTensor()\n",
|
|
")\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# Create image size (from Table 3 in the ViT paper) \n",
|
|
"IMG_SIZE = 224\n",
|
|
"\n",
|
|
"# Create transform pipeline manually\n",
|
|
"manual_transforms = transforms.Compose([\n",
|
|
" transforms.Resize((IMG_SIZE, IMG_SIZE)),\n",
|
|
" transforms.ToTensor(),\n",
|
|
"]) \n",
|
|
"print(f\"Manually created transforms: {manual_transforms}\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 8,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/"
|
|
},
|
|
"id": "W4vWgIprDJau",
|
|
"outputId": "6de4b52b-1570-408a-8399-8f6465258a31"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"(<torch.utils.data.dataloader.DataLoader at 0x7f7e08f76c90>,\n",
|
|
" <torch.utils.data.dataloader.DataLoader at 0x7f7e0889f590>,\n",
|
|
" ['pizza', 'steak', 'sushi'])"
|
|
]
|
|
},
|
|
"execution_count": 8,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"# Set the batch size\n",
|
|
"BATCH_SIZE = 32 # this is lower than the ViT paper but it's because we're starting small\n",
|
|
"\n",
|
|
"# Create data loaders\n",
|
|
"train_dataloader, test_dataloader, class_names = data_setup.create_dataloaders(\n",
|
|
" train_dir=train_dir,\n",
|
|
" test_dir=test_dir,\n",
|
|
" transform=manual_transforms, # use manually created transforms\n",
|
|
" batch_size=BATCH_SIZE\n",
|
|
")\n",
|
|
"\n",
|
|
"train_dataloader, test_dataloader, class_names"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 9,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/"
|
|
},
|
|
"id": "u7eLIFHyDJRr",
|
|
"outputId": "464ae045-dddd-40a4-ef6c-3414e3ebc787"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"(torch.Size([3, 224, 224]), tensor(0))"
|
|
]
|
|
},
|
|
"execution_count": 9,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"# Get a batch of images\n",
|
|
"image_batch, label_batch = next(iter(train_dataloader))\n",
|
|
"\n",
|
|
"# Get a single image from the batch\n",
|
|
"image, label = image_batch[0], label_batch[0]\n",
|
|
"\n",
|
|
"# View the batch shapes\n",
|
|
"image.shape, label"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 10,
|
|
"metadata": {
|
|
"colab": {
|
|
"base_uri": "https://localhost:8080/",
|
|
"height": 264
|
|
},
|
|
"id": "2yyNHCmCDbSR",
|
|
"outputId": "da4b3c3a-6b9a-4954-9b44-226c89f89a8e"
|
|
},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 432x288 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {
|
|
"needs_background": "light"
|
|
},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"# Plot image with matplotlib\n",
|
|
"plt.imshow(image.permute(1, 2, 0)) # rearrange image dimensions to suit matplotlib [color_channels, height, width] -> [height, width, color_channels]\n",
|
|
"plt.title(class_names[label])\n",
|
|
"plt.axis(False);"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "nwmoMhW8IqSu"
|
|
},
|
|
"source": [
|
|
"## 1. Replicate the ViT architecture we created with in-built [PyTorch transformer layers](https://pytorch.org/docs/stable/nn.html#transformer-layers).\n",
|
|
"\n",
|
|
"* You'll want to look into replacing our `TransformerEncoderBlock()` class with [`torch.nn.TransformerEncoderLayer()`](https://pytorch.org/docs/stable/generated/torch.nn.TransformerEncoderLayer.html#torch.nn.TransformerEncoderLayer) (these contain the same layers as our custom blocks). \n",
|
|
"* You can stack `torch.nn.TransformerEncoderLayer()`'s on top of each other with [`torch.nn.TransformerEncoder()`](https://pytorch.org/docs/stable/generated/torch.nn.TransformerEncoder.html#torch.nn.TransformerEncoder)."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 11,
|
|
"metadata": {
|
|
"id": "pmDd_YZ7VSrL"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# TODO: your code"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "MBWnDZao9w_5"
|
|
},
|
|
"source": [
|
|
"## 2. Turn the custom ViT architecture we created into a Python script, for example, `vit.py`.\n",
|
|
"\n",
|
|
"* You should be able to import an entire ViT model using something like`from vit import ViT`.\n",
|
|
"* We covered the art of turning code cells into Python scrips in [05. PyTorch Going Modular](https://www.learnpytorch.io/05_pytorch_going_modular/). \n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 12,
|
|
"metadata": {
|
|
"id": "NFXVZNCzVYgV"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# TODO: your code"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "aTKbje-e9118"
|
|
},
|
|
"source": [
|
|
"## 3. Train a pretrained ViT feature extractor model (like the one we made in [08. PyTorch Paper Replicating section 10](https://www.learnpytorch.io/08_pytorch_paper_replicating/#10-bring-in-pretrained-vit-from-torchvisionmodels-on-same-dataset)) on 20% of the pizza, steak and sushi data like the dataset we used in [07. PyTorch Experiment Tracking section 7.3](https://www.learnpytorch.io/07_pytorch_experiment_tracking/#73-download-different-datasets) \n",
|
|
"* See how it performs compared to the EffNetB2 model we compared it to in [08. PyTorch Paper Replicating section 10.6](https://www.learnpytorch.io/08_pytorch_paper_replicating/#106-save-feature-extractor-vit-model-and-check-file-size)."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 13,
|
|
"metadata": {
|
|
"id": "R7iKYRAUVkA7"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# TODO: your code"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "LH-vHr3m9_oH"
|
|
},
|
|
"source": [
|
|
"## 4. Try repeating the steps from excercise 3 but this time use the \"`ViT_B_16_Weights.IMAGENET1K_SWAG_E2E_V1`\" pretrained weights from [`torchvision.models.vit_b_16()`](https://pytorch.org/vision/stable/models/generated/torchvision.models.vit_b_16.html#torchvision.models.vit_b_16).\n",
|
|
"* Note: ViT pretrained with SWAG weights has a minimum input image size of (384, 384), though this is accessible in the weights `.transforms()` method."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 14,
|
|
"metadata": {
|
|
"id": "dWxceTz3VmeB"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# TODO: your code"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {
|
|
"id": "ZLcCgRhS-OhV"
|
|
},
|
|
"source": [
|
|
"# 5. Our custom ViT model architecture closely mimics that of the ViT paper, however, our training recipe misses a few things. \n",
|
|
"* Research some of the following topics from Table 3 in the ViT paper that we miss and write a sentence about each and how it might help with training:\n",
|
|
" * **ImageNet-21k pretraining** \n",
|
|
" * **Learning rate warmup** \n",
|
|
" * **Learning rate decay** \n",
|
|
" * **Gradient clipping** "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 15,
|
|
"metadata": {
|
|
"id": "7HwonCSsVtnr"
|
|
},
|
|
"outputs": [],
|
|
"source": [
|
|
"# TODO: your explanations of the above terms"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"accelerator": "GPU",
|
|
"colab": {
|
|
"authorship_tag": "ABX9TyOhoCjGZZxrecbm76R8UJZn",
|
|
"collapsed_sections": [],
|
|
"include_colab_link": true,
|
|
"name": "08_pytorch_paper_replicating_exercises.ipynb",
|
|
"provenance": [],
|
|
"toc_visible": true
|
|
},
|
|
"gpuClass": "standard",
|
|
"kernelspec": {
|
|
"display_name": "Python 3",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"name": "python"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 0
|
|
}
|