diff --git "a/tutorials/load_data.ipynb" "b/tutorials/load_data.ipynb"
new file mode 100644--- /dev/null
+++ "b/tutorials/load_data.ipynb"
@@ -0,0 +1,7793 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "7QWi2mZTNM0q"
+ },
+ "source": [
+ "# Loading the SPIDER dataset from HuggingFace\n",
+ "\n",
+ "This tutorial will walk you through the steps to download and use the SPIDER dataset."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "L9f_x4IO9pka"
+ },
+ "source": [
+ "### Table of Contents\n",
+ "\n",
+ "1. [Installing Dependencies](#dependencies)\n",
+ "2. [Loading Demo Dataset](#demo_config)\n",
+ "3. [Visualizing an Example Image](#visualizing_image)\n",
+ "4. [Resizing Images](#resizing)\n",
+ "5. [Loading Original Images](#original_images)\n",
+ "6. [Extracting Metadata](#metadata)\n",
+ "7. [Filtering Scan Types](#filter_scan_type)\n",
+ "8. [Loading Full Dataset](#loading_full_dataset)\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "Wt9vqFfcQT8G"
+ },
+ "source": [
+ "### Installing Dependencies "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "ejAzA9RlNSVv"
+ },
+ "source": [
+ "First, install the necessary dependencies:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {
+ "id": "cRMYzjBvo66Q",
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "outputId": "e1432cce-dda3-4905-dcd3-9781a2bbdfc8"
+ },
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m510.5/510.5 kB\u001b[0m \u001b[31m5.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m116.3/116.3 kB\u001b[0m \u001b[31m7.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m194.1/194.1 kB\u001b[0m \u001b[31m6.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m134.8/134.8 kB\u001b[0m \u001b[31m8.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m52.7/52.7 MB\u001b[0m \u001b[31m25.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
+ "\u001b[?25h"
+ ]
+ }
+ ],
+ "source": [
+ "!pip install datasets -q\n",
+ "!pip install scikit-image -q\n",
+ "!pip install SimpleITK -q"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "LOBqCjn3MDm_",
+ "outputId": "a1ade988-ccb4-40ea-9038-faa6792325d5"
+ },
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "datasets: 2.18.0\n",
+ "scikit-image: 0.19.3\n",
+ "SimpleITK: 2.3.1\n"
+ ]
+ }
+ ],
+ "source": [
+ "import datasets\n",
+ "import skimage\n",
+ "import SimpleITK as sitk\n",
+ "\n",
+ "print(f'datasets: {datasets.__version__}')\n",
+ "print(f'scikit-image: {skimage.__version__}')\n",
+ "print(f'SimpleITK: {sitk.__version__}')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "Liq722klQal4"
+ },
+ "source": [
+ "### Loading the Dataset with Demo Configuration "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "zHCVOF9xOZFX"
+ },
+ "source": [
+ "Next, use the `load_dataset` function from the `datasets` library to download the data directly from the [Zenodo](https://zenodo.org/records/10159290) repository.\n",
+ "\n",
+ "Select the `demo` configuration to verify that the function works as intended. The `demo` configuration downloads all of the original `.mha` image and mask files from Zenodo, but only processes the first 10 examples to reduce computation time. The downloaded `.mha` image files will be saved to cache on your local system (which you can set with the `cache_dir` parameter -- see the HuggingFace [docs](https://huggingface.co/docs/datasets/v2.18.0/en/package_reference/loading_methods#datasets.load_dataset)).\n",
+ "\n",
+ "Note that in future versions of the `load_dataset` function, you will have to\n",
+ "explicitly pass `trust_remote_code=True` for the code to run. You can review\n",
+ "the source code in the HuggingFace repository [here](https://huggingface.co/datasets/cdoswald/SPIDER/blob/main/SPIDER.py)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 441,
+ "referenced_widgets": [
+ "621343e4760c43f0ab55d3ba8415a43f",
+ "250923e24b104d80bae010903ae7886e",
+ "c94edf1892dc4a75926b4fcdc72538ec",
+ "e1a59a81e3f142d5bb979ab9fd60e25d",
+ "815b0613252547c5ba4bfb5d43f6bf5c",
+ "48c41139f5494157988be8c1a846c8bd",
+ "5fae5e920e3440e4b65313d2d4c0b421",
+ "723499d614344423a087580bdc64c6c4",
+ "051bb482f41b40d9af6a9b21090080b5",
+ "b82b605406584f3492f217b34f75dd47",
+ "23423b6a9ff84610b1110e2106fc00a8",
+ "f02765b1804545b4b4acec881578d49a",
+ "5f4a20ed7ff44afd90e95e2e3b31fdd2",
+ "8be26f5759fa4ad9b5dfcf7051d1c1f5",
+ "87d3bdcde951426aba06009e8bdd510f",
+ "bbea68cb4dab4f5195c41021a34c571f",
+ "acfc87ff0d7c4cbcaa4cfa91516144d2",
+ "a791255e8ee04ba4b14f111e0ed7719d",
+ "7b6807645f6e48ca9d2cc46a8ff3e611",
+ "267b26ee9864470b9917568305f63541",
+ "d1c0f001e1d644b09e049faeb84c3f53",
+ "6e2d23576f3f4df6ba5b254f8045901f",
+ "3d58235d864545a6b8d18af6dc467add",
+ "4ea0a5051a7240fe8f6bfd61ac88b656",
+ "a2d0a6d375b74f06a9f5d12403b88816",
+ "410df446d3754fd8b5531c85c6809470",
+ "650474a3bd254c4286cc6d537bcfd62e",
+ "af10fca1f7b1483eb2ef8bf1179c6cae",
+ "15534a65241646518315ad67cc97c08e",
+ "3d9988ce8c114b8b878bad95cc06a458",
+ "9a73f08d0245457397e817cbf247f779",
+ "b946c44b275640eca910a4a208e4562e",
+ "db70ea21aa944583ad4a363de82bb2d5",
+ "cd207b34f4d148a688d3deda6571c80d",
+ "c04510a1418744f5a18447f87c675204",
+ "0edd7b2cb75446a1a0c96d3e8fbe431c",
+ "63a0a86d2fab478f8ed520289313ab90",
+ "aaa88cf1e23d4d328749281cf82682f6",
+ "9fcfc169527041d79c8bdc660413d6e0",
+ "79d4513c94ff44aba803333304c36a69",
+ "d60f493361b642a4bbcc3b9b54443010",
+ "1567b4c0cafe47a9a7510584558a913b",
+ "d7a5387e195d47e29576d4c6d29dcb2b",
+ "36580c8feaed4fe085e8aaa5191d9388",
+ "225f0f74a40c4105b16feee76e0f118c",
+ "0272e5ff7cf84580a7b88e25779024ed",
+ "eb6c72632aa84bb29c1db28ddb226af4",
+ "69a63fa0fecc470fab787a11b17982ab",
+ "86103848746442788d56fdd06e48c798",
+ "60951750f0d94eed8ab8ac3677a694d3",
+ "2c40b2fba8cd4a88aa8d4ffe9e6ee22e",
+ "820211f82fc046b4b647c64fc181affb",
+ "85736668956245eaabe456c91ff298c9",
+ "ccfe041d29c948fa9c9ba28de22fa205",
+ "dfb6ab1c1532469aaafc24549e73fa76",
+ "f598b908c9344a2d826a526b283699d3",
+ "7b3aaf29306b4050b5db89962401439e",
+ "cc8e2735483c4d2abb5412ba18b5c2a9",
+ "57ec61b6d54e4779ab007c165a47db05",
+ "cf576920a45c419bbe5b18d9eb7b8a0f",
+ "32423c538fe84a1abc8f78b7d3a2590e",
+ "ed43c1c77b364862856c9dcd0b841389",
+ "b3e982dab99d4ae68d970d76641fb994",
+ "0e3214c3ecdc44a4be5c1fd2bf41af3f",
+ "41e3206455614968af6bfdc6933b46b9",
+ "aaee835fd5534b0b8ae59b02822a2964",
+ "528ea2f5936240158bda15a9b8f746cf",
+ "fbef287ed7aa4ec98ec51ad7a141b5c6",
+ "6ad4cbf87ea54292ad66eb036885d75b",
+ "e07de8941e974c4b94b6566c91cac95b",
+ "a8f7426d733a46378d6472cae8c6675d",
+ "49c5a91d6e69463ab1b70275bf874f04",
+ "5cd82c15589f46e1964b2769266f9b51",
+ "ecce67a0cecd445c8a5af2138404978c",
+ "006b58baedc04f8d9ded64efd69e577d",
+ "eadefca20d5548d097e56f7db73eb830",
+ "1dc698cb39404b29a05ab69b2aeaa9a2",
+ "4391604f31854468b3edd9a0a113ee49",
+ "1984044ab86c472a8b2e6eb975a3b793",
+ "6d4944cce17340198b7aea42fba085f8",
+ "d146bd18851c4c1eaf90051fa9e5965c",
+ "8eaa07b051a8434fb8f42ae1ab04eaac",
+ "4522219e9220461cb530960a24fdd9d6",
+ "090afa3384734f1ab40a6578cc780c92",
+ "c4f68c215dfe4165971806aa64e845a7",
+ "6c3d708200ea4d89a6e1895d551d932d",
+ "e5a03ceac7f7493e8baa1508c2ba102c",
+ "9344b1f91aaa48e7b31a23dca3c0b39e",
+ "95a4bbe5b62a46fe82775be9d3cff3b0",
+ "deb4ed22013e4f4583c1805f7d2da24f",
+ "3e2e261dc5f641ddb84c864e0e6d97da",
+ "eb00e7ad026242b38b952db8096ec026",
+ "6f0185f0b64d45baace00fd0755e71f5",
+ "5f2f4a9cd5914a52b5f6bd154365771b",
+ "6fa9005edc854dc6b3c48afc8b49bd3d",
+ "cdf035c3e4154435a8daa39f5e48e8c6",
+ "1aeb813172f84cd285d329eb072ca699",
+ "48dc7a64bfa74727bb24bfff56d23a15",
+ "c73511bc40c64286940563fe06c77740",
+ "627ca8e891ab4fb6be5893c6dffbe6a0",
+ "76583c6c2c0f4de3883a339667831cb1",
+ "df5fd312a6ef40649ffedba0fb1573cc",
+ "bd7e885e3e26492bbdb8077e9f320392",
+ "8e1f792cb9824d3496cafefc9d248c10",
+ "14fc73fbc9dd41248da4c1289f0c8d6c",
+ "a99ace5b8ffc407fb64693f2ee07d59e",
+ "c1d7ecb478894f04b3aab0197f9eb58b",
+ "4139fd0e5894424795628262968fa272",
+ "76bdff969cd04ad0a036dc8a637a683f",
+ "fa22c564cdee48a3911ca65243d8b074"
+ ]
+ },
+ "id": "QOHzp7bRoqV4",
+ "outputId": "2bdc567d-6ee6-4ad2-cec7-5997bbfb1880"
+ },
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stderr",
+ "text": [
+ "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning: \n",
+ "The secret `HF_TOKEN` does not exist in your Colab secrets.\n",
+ "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n",
+ "You will be able to reuse this secret in all of your notebooks.\n",
+ "Please note that authentication is recommended but still optional to access public models or datasets.\n",
+ " warnings.warn(\n"
+ ]
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading builder script: 0%| | 0.00/20.5k [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "621343e4760c43f0ab55d3ba8415a43f"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading readme: 0%| | 0.00/6.23k [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "f02765b1804545b4b4acec881578d49a"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading data: 0%| | 0.00/3.70G [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "3d58235d864545a6b8d18af6dc467add"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading data: 0%| | 0.00/58.2M [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "cd207b34f4d148a688d3deda6571c80d"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading data: 0.00B [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "225f0f74a40c4105b16feee76e0f118c"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading data: 0.00B [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "f598b908c9344a2d826a526b283699d3"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Downloading data: 0%| | 0.00/1.20k [00:00, ?B/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "528ea2f5936240158bda15a9b8f746cf"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Generating train split: 0 examples [00:00, ? examples/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "4391604f31854468b3edd9a0a113ee49"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Generating validation split: 0 examples [00:00, ? examples/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "95a4bbe5b62a46fe82775be9d3cff3b0"
+ }
+ },
+ "metadata": {}
+ },
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "Generating test split: 0 examples [00:00, ? examples/s]"
+ ],
+ "application/vnd.jupyter.widget-view+json": {
+ "version_major": 2,
+ "version_minor": 0,
+ "model_id": "627ca8e891ab4fb6be5893c6dffbe6a0"
+ }
+ },
+ "metadata": {}
+ }
+ ],
+ "source": [
+ "from datasets import load_dataset\n",
+ "\n",
+ "dataset = load_dataset(\"cdoswald/SPIDER\", name=\"demo\", trust_remote_code=True)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "W0ODUGtORTZ6"
+ },
+ "source": [
+ "Notice that the dataset is split into train, validation, and test subsets, and each example has 8 features: `patient_id`, `scan_type`, `image`, `mask`, `image_path`, `mask_path`, `metadata`, and `rad_gradings`. Only the first 10 examples were processed for this `demo` configuration."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "2G6WIkFZb9Up",
+ "outputId": "702c9809-5398-4584-8ae8-5e35b2b2700c"
+ },
+ "outputs": [
+ {
+ "output_type": "execute_result",
+ "data": {
+ "text/plain": [
+ "DatasetDict({\n",
+ " train: Dataset({\n",
+ " features: ['patient_id', 'scan_type', 'image', 'mask', 'image_path', 'mask_path', 'metadata', 'rad_gradings'],\n",
+ " num_rows: 10\n",
+ " })\n",
+ " validation: Dataset({\n",
+ " features: ['patient_id', 'scan_type', 'image', 'mask', 'image_path', 'mask_path', 'metadata', 'rad_gradings'],\n",
+ " num_rows: 10\n",
+ " })\n",
+ " test: Dataset({\n",
+ " features: ['patient_id', 'scan_type', 'image', 'mask', 'image_path', 'mask_path', 'metadata', 'rad_gradings'],\n",
+ " num_rows: 10\n",
+ " })\n",
+ "})"
+ ]
+ },
+ "metadata": {},
+ "execution_count": 4
+ }
+ ],
+ "source": [
+ "dataset"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "HLX45XUvUDKs"
+ },
+ "source": [
+ "### Visualizing an Example Image "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "34CeKbtlSH3m"
+ },
+ "source": [
+ "We can view the features for a specific example by first selecting the data subset (e.g., \"train\") and then indexing a particular observation:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/"
+ },
+ "id": "aTPm0k0vSSDB",
+ "outputId": "a18bb90d-17fb-4ec2-c133-f578f056dc84"
+ },
+ "outputs": [
+ {
+ "output_type": "stream",
+ "name": "stdout",
+ "text": [
+ "Patient ID: 184\n",
+ "Scan type: t1\n"
+ ]
+ }
+ ],
+ "source": [
+ "example = dataset['train'][0]\n",
+ "print(f'Patient ID: {example[\"patient_id\"]}\\nScan type: {example[\"scan_type\"]}')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {
+ "id": "Y4hgVOivSwf4"
+ },
+ "source": [
+ "By default, the `image` and `mask` attributes will each be 3D volumetric arrays of size `(512, 512, 30)` -- in other words, 30 stacked 512 x 512 grayscale images (note that the channel dimension indicates depth rather than RGB values).\n",
+ "\n",
+ "We can select a few specific depths to display as 2D images using matplotlib's `imshow` function:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {
+ "colab": {
+ "base_uri": "https://localhost:8080/",
+ "height": 356
+ },
+ "id": "Nq0jUm-UazpL",
+ "outputId": "acb70db5-f568-4aaf-e0c5-ca4473d0039b"
+ },
+ "outputs": [
+ {
+ "output_type": "display_data",
+ "data": {
+ "text/plain": [
+ "