careamics 0.0.19__py3-none-any.whl

careamics/dataset_ng/demos/demo_datamodule.ipynb

@@ -0,0 +1,447 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from careamics.config.configuration_factories import (\n",
+    "    _create_ng_data_configuration,\n",
+    "    _list_spatial_augmentations,\n",
+    ")\n",
+    "from careamics.lightning.dataset_ng.data_module import CareamicsDataModule"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%load_ext autoreload\n",
+    "%autoreload 2\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pathlib import Path\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import tifffile\n",
+    "from careamics_portfolio import PortfolioManager\n",
+    "\n",
+    "# instantiate data portfolio manage\n",
+    "portfolio = PortfolioManager()\n",
+    "\n",
+    "# and download the data\n",
+    "root_path = Path(\"./data\")\n",
+    "files = portfolio.denoising.N2V_BSD68.download(root_path)\n",
+    "\n",
+    "# create paths for the data\n",
+    "data_path = Path(root_path / \"denoising-N2V_BSD68.unzip/BSD68_reproducibility_data\")\n",
+    "train_path = data_path / \"train\"\n",
+    "val_path = data_path / \"val\"\n",
+    "test_path = data_path / \"test\" / \"images\"\n",
+    "gt_path = data_path / \"test\" / \"gt\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "image_std, image_mean = [], []\n",
+    "for file in train_path.glob(\"*.tiff\"):\n",
+    "    image = tifffile.imread(file)\n",
+    "    image_std.append(image.std())\n",
+    "    image_mean.append(image.mean())\n",
+    "image_std, image_mean = np.mean(image_std), np.mean(image_mean)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# from path, train and val, no target\n",
+    "\n",
+    "config = _create_ng_data_configuration(\n",
+    "    data_type=\"tiff\",\n",
+    "    axes=\"SYX\",\n",
+    "    patch_size=(64, 64),\n",
+    "    batch_size=64,\n",
+    "    augmentations=_list_spatial_augmentations()\n",
+    ")\n",
+    "\n",
+    "config.set_means_and_stds([image_mean], [image_std])\n",
+    "config.val_dataloader_params = {\"shuffle\": False}\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    train_data=train_path,\n",
+    "    val_data=val_path,\n",
+    ")\n",
+    "data_module.setup('fit')\n",
+    "data_module.setup('validate')\n",
+    "\n",
+    "train_batch = next(iter(data_module.train_dataloader()))\n",
+    "val_batch = next(iter(data_module.val_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(train_batch[0].data[i][0].numpy(), cmap=\"gray\")\n",
+    "\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(val_batch[0].data[i][0].numpy(), cmap=\"gray\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# from path, only predict\n",
+    "\n",
+    "from careamics.config.data import NGDataConfig\n",
+    "\n",
+    "config = NGDataConfig(\n",
+    "    data_type=\"tiff\",\n",
+    "    patching={\n",
+    "        \"name\": \"tiled\",\n",
+    "        \"patch_size\": (128, 128),\n",
+    "        \"overlaps\": (32, 32)\n",
+    "    },\n",
+    "    axes=\"YX\",\n",
+    "    batch_size=8,\n",
+    "    image_means=[image_mean],\n",
+    "    image_stds=[image_std]\n",
+    ")\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    pred_data=test_path\n",
+    ")\n",
+    "data_module.setup('predict')\n",
+    "\n",
+    "pred_batch = next(iter(data_module.predict_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(pred_batch[0].data[i][0].numpy(), cmap=\"gray\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# test from array"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train_array = tifffile.imread(sorted(train_path.rglob('*'))[0])\n",
+    "val_array = tifffile.imread(sorted(val_path.rglob('*'))[0])\n",
+    "test_array = tifffile.imread(sorted(test_path.rglob('*'))[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# from array, train and val, no target\n",
+    "\n",
+    "config = _create_ng_data_configuration(\n",
+    "    data_type=\"array\",\n",
+    "    axes=\"SYX\",\n",
+    "    patch_size=(64, 64),\n",
+    "    batch_size=64,\n",
+    "    augmentations=_list_spatial_augmentations()\n",
+    ")\n",
+    "\n",
+    "config.set_means_and_stds([image_mean], [image_std])\n",
+    "config.val_dataloader_params = {\"shuffle\": False}\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    train_data=train_array,\n",
+    "    val_data=val_array,\n",
+    ")\n",
+    "data_module.setup('fit')\n",
+    "data_module.setup('validate')\n",
+    "\n",
+    "train_batch = next(iter(data_module.train_dataloader()))\n",
+    "val_batch = next(iter(data_module.val_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(train_batch[0].data[i][0].numpy(), cmap=\"gray\")\n",
+    "\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(val_batch[0].data[i][0].numpy(), cmap=\"gray\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# test with target"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import skimage\n",
+    "\n",
+    "example_data = skimage.data.human_mitosis()\n",
+    "\n",
+    "markers = np.zeros_like(example_data)\n",
+    "markers[example_data < 25] = 1\n",
+    "markers[example_data > 50] = 2\n",
+    "\n",
+    "elevation_map = skimage.filters.sobel(example_data)\n",
+    "segmentation = skimage.segmentation.watershed(elevation_map, markers)\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 2)\n",
+    "ax[0].imshow(example_data)\n",
+    "ax[1].imshow(segmentation)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "config = _create_ng_data_configuration(\n",
+    "    data_type=\"array\",\n",
+    "    axes=\"YX\",\n",
+    "    patch_size=(64, 64),\n",
+    "    batch_size=64,\n",
+    "    augmentations=_list_spatial_augmentations()\n",
+    ")\n",
+    "config.set_means_and_stds(\n",
+    "    [example_data.mean()],\n",
+    "    [example_data.std()],\n",
+    "    [segmentation.mean()],\n",
+    "    [segmentation.std()]\n",
+    ")\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    train_data=[example_data],\n",
+    "    train_data_target=[segmentation],\n",
+    "    val_data=[example_data],\n",
+    "    val_data_target=[segmentation]\n",
+    ")\n",
+    "data_module.setup('fit')\n",
+    "data_module.setup('validate')\n",
+    "\n",
+    "train_batch = next(iter(data_module.train_dataloader()))\n",
+    "val_batch = next(iter(data_module.val_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(2, 8, figsize=(10, 3))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[0][i].imshow(train_batch[0].data[i][0].numpy(), cmap=\"gray\")\n",
+    "    ax[1][i].imshow(train_batch[1].data[i][0].numpy())\n",
+    "\n",
+    "\n",
+    "fig, ax = plt.subplots(2, 8, figsize=(10, 3))\n",
+    "for i in range(8):\n",
+    "    ax[0][i].imshow(val_batch[0].data[i][0].numpy(), cmap=\"gray\")\n",
+    "    ax[1][i].imshow(val_batch[1].data[i][0].numpy())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# from array, only predict, with target\n",
+    "\n",
+    "from careamics.config.data import NGDataConfig\n",
+    "\n",
+    "config = NGDataConfig(\n",
+    "    data_type=\"array\",\n",
+    "    patching={\n",
+    "        \"name\": \"tiled\",\n",
+    "        \"patch_size\": (128, 128),\n",
+    "        \"overlaps\": (32, 32)\n",
+    "    },\n",
+    "    axes=\"YX\",\n",
+    "    batch_size=8,\n",
+    "    image_means=[image_mean],\n",
+    "    image_stds=[image_std]\n",
+    ")\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    pred_data=example_data,\n",
+    "    pred_data_target=segmentation\n",
+    ")\n",
+    "data_module.setup('predict')\n",
+    "\n",
+    "pred_batch = next(iter(data_module.predict_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(pred_batch[0].data[i][0].numpy(), cmap=\"gray\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# from list of paths"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "config = _create_ng_data_configuration(\n",
+    "    data_type=\"tiff\",\n",
+    "    axes=\"SYX\",\n",
+    "    patch_size=(64, 64),\n",
+    "    batch_size=64,\n",
+    "    augmentations=_list_spatial_augmentations()\n",
+    ")\n",
+    "\n",
+    "config.set_means_and_stds([image_mean], [image_std])\n",
+    "config.val_dataloader_params = {\"shuffle\": False}\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    train_data=sorted(train_path.glob(\"*.tiff\")),\n",
+    "    val_data=sorted(val_path.glob(\"*.tiff\")),\n",
+    ")\n",
+    "data_module.setup('fit')\n",
+    "data_module.setup('validate')\n",
+    "\n",
+    "train_batch = next(iter(data_module.train_dataloader()))\n",
+    "val_batch = next(iter(data_module.val_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(train_batch[0].data[i][0].numpy(), cmap=\"gray\")\n",
+    "\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(val_batch[0].data[i][0].numpy(), cmap=\"gray\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# from custom"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "config = _create_ng_data_configuration(\n",
+    "    data_type=\"custom\",\n",
+    "    axes=\"SYX\",\n",
+    "    patch_size=(64, 64),\n",
+    "    batch_size=64,\n",
+    "    augmentations=_list_spatial_augmentations()\n",
+    ")\n",
+    "\n",
+    "config.set_means_and_stds([image_mean], [image_std])\n",
+    "config.val_dataloader_params = {\"shuffle\": False}\n",
+    "\n",
+    "def read_source_func(path):\n",
+    "    image = tifffile.imread(path)\n",
+    "    image = 255 - image\n",
+    "    return image\n",
+    "\n",
+    "data_module = CareamicsDataModule(\n",
+    "    data_config=config,\n",
+    "    train_data=sorted(train_path.glob(\"*.tiff\")),\n",
+    "    val_data=sorted(val_path.glob(\"*.tiff\")),\n",
+    "    read_source_func=read_source_func\n",
+    ")\n",
+    "data_module.setup('fit')\n",
+    "data_module.setup('validate')\n",
+    "\n",
+    "train_batch = next(iter(data_module.train_dataloader()))\n",
+    "val_batch = next(iter(data_module.val_dataloader()))\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(train_batch[0].data[i][0].numpy(), cmap=\"gray\")\n",
+    "\n",
+    "\n",
+    "fig, ax = plt.subplots(1, 8, figsize=(10, 5))\n",
+    "for i in range(8):\n",
+    "    ax[i].imshow(val_batch[0].data[i][0].numpy(), cmap=\"gray\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "czi",
+   "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": "3.12.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}