custom dataset

hervevillard · Sep 11, 2024 · 774181e · 774181e
1 parent ba9c505
commit 774181e
Show file tree

Hide file tree

Showing 3 changed files with 585 additions and 5 deletions.
diff --git a/03_pytorch_computer_vision.ipynb b/03_pytorch_computer_vision.ipynb
diff --git a/04_custom_data_creation.ipynb b/04_custom_data_creation.ipynb
@@ -0,0 +1,229 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# PyTorch Custom Data Creation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch\n",
+    "import torchvision\n",
+    "import torchvision.datasets as datasets \n",
+    "import torchvision.transforms as transforms\n",
+    "\n",
+    "# setup directory\n",
+    "import pathlib\n",
+    "data_dir = pathlib.Path(\"../data\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "### Download data\n",
+    "\n",
+    "Get the Food101 dataset from PyTorch.\n",
+    "\n",
+    "Food101 in torchvision.datasets - https://pytorch.org/vision/stable/generated/torchvision.datasets.Food101.html\n",
+    "Original Food101 dataset - https://data.vision.ee.ethz.ch/cvl/datasets_extra/food-101/\n",
+    "\n",
+    "Note: Downloading the dataset from PyTorch may take ~10-15 minutes depending on your internet speed. It will download ~5GB of data to the specified root directory.\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Downloading https://data.vision.ee.ethz.ch/cvl/food-101.tar.gz to ..\\data\\food-101.tar.gz\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      " 47%|████▋     | 2346024960/4996278331 [03:34<04:53, 9022497.18it/s] "
+     ]
+    }
+   ],
+   "source": [
+    "# Get training data\n",
+    "train_data = datasets.Food101(root=data_dir,\n",
+    "                              split=\"train\",\n",
+    "                              #transform=transforms.ToTensor(),\n",
+    "                              download=True)\n",
+    "# Get testing \n",
+    "test_data =  datasets.Food101(root=data_dir,\n",
+    "                              split=\"test\",\n",
+    "                              #transform=transforms.ToTensor(),\n",
+    "                              download=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class_names = train_data.classes\n",
+    "class_names[:10]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# View first sample (PIL image format)\n",
+    "print(class_names[train_data[0][1]])\n",
+    "train_data[0][0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "### Find subset of appropriate classes\n",
+    "\n",
+    "Want: Steak, pizza, sushi.\n",
+    "\n",
+    "Current path setup:\n",
+    "\n",
+    "../data/food-101/images/CLASS_NAME/IMAGES.jpg\n",
+    "\n",
+    "Going to get a list of the different target image classes (pizza, steak, sushi) filenames and then copy the images to separate folders.\n",
+    "\n",
+    "I'd like to get a random 10% of the images from the target classes from both datasets.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "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": "3.12.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}