1.MONAI DDPM with MedNist.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e9d82774-2fa5-4349-9e75-6dd49a6c5169",
   "metadata": {},
   "source": [
    "# ISBI 2024 DiMEDIA: MONAI Tutorial: DDPM with MedNIST Dataset\n",
    "\n",
    "MONAI is a PyTorch-based, open-source framework for deep learning in healthcare imaging, part of PyTorch Ecosystem. Its ambitions are:\n",
    "\n",
    "- developing a community of academic, industrial and clinical researchers collaborating on a common foundation;\n",
    "- creating state-of-the-art, end-to-end training workflows for healthcare imaging;\n",
    "- providing researchers with the optimized and standardized way to create and evaluate deep learning models.\n",
    "\n",
    "## Get Started\n",
    "\n",
    "MONAI contains three main components: MONAI Core, MONAI Label and MONAI Deploy. For tutorial use, we only adopt the MONAI Core and MONAI Label. \n",
    "\n",
    "To install the current release, run the following command:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "253b18d8-e5aa-4d84-ba44-c783bc4edd14",
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install -q matplotlib monai monailabel monai-generative\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60f8bd5b-8884-488e-9bea-945dc5fc5990",
   "metadata": {},
   "source": [
    "First impor the necessary packages, and you can verify the installation by:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "18276971-9548-4f09-a122-3053e0a56830",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MONAI version: 1.3.0\n",
      "Numpy version: 1.26.2\n",
      "Pytorch version: 2.2.2+cu118\n",
      "MONAI flags: HAS_EXT = False, USE_COMPILED = False, USE_META_DICT = False\n",
      "MONAI rev id: 865972f7a791bf7b42efbcd87c8402bd865b329e\n",
      "MONAI __file__: /remote/rds/users/<username>/miniforge3/envs/dev/lib/python3.10/site-packages/monai/__init__.py\n",
      "\n",
      "Optional dependencies:\n",
      "Pytorch Ignite version: 0.4.11\n",
      "ITK version: 5.3.0\n",
      "Nibabel version: 5.2.1\n",
      "scikit-image version: 0.22.0\n",
      "scipy version: 1.11.4\n",
      "Pillow version: 10.3.0\n",
      "Tensorboard version: 2.16.2\n",
      "gdown version: 4.7.3\n",
      "TorchVision version: 0.17.2+cu118\n",
      "tqdm version: 4.66.1\n",
      "lmdb version: 1.4.1\n",
      "psutil version: 5.9.5\n",
      "pandas version: 2.1.4\n",
      "einops version: 0.7.0\n",
      "transformers version: 4.39.3\n",
      "mlflow version: 2.11.3\n",
      "pynrrd version: 1.0.0\n",
      "clearml version: NOT INSTALLED or UNKNOWN VERSION.\n",
      "\n",
      "For details about installing the optional dependencies, please visit:\n",
      "    https://docs.monai.io/en/latest/installation.html#installing-the-recommended-dependencies\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import shutil\n",
    "import tempfile\n",
    "import time\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn.functional as F\n",
    "import monai\n",
    "from monai import transforms\n",
    "from monai.apps import MedNISTDataset\n",
    "from monai.config import print_config\n",
    "from monai.data import CacheDataset, DataLoader\n",
    "from monai.utils import first, set_determinism\n",
    "from torch.cuda.amp import GradScaler, autocast\n",
    "from tqdm import tqdm\n",
    "\n",
    "from generative.inferers import DiffusionInferer\n",
    "from generative.networks.nets import DiffusionModelUNet\n",
    "from generative.networks.schedulers import DDPMScheduler\n",
    "\n",
    "set_determinism(42)\n",
    "\n",
    "monai.config.print_config()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fdb059ac-bf02-4d25-af1b-4c78313ff9ab",
   "metadata": {},
   "source": [
    "## Set up the MedNIST Dataset\n",
    "\n",
    "In this tutorial, we will train our models on the MedNIST dataset available on MONAI (https://docs.monai.io/en/stable/apps.html#monai.apps.MedNISTDataset). \n",
    "\n",
    "In order to train faster, we will select just one of the available classes (\"Hand\"), resulting in a training set with 7999 2D images.\n",
    "\n",
    "You can specify a directory with the MONAI_DATA_DIRECTORY environment variable. For the convenience, we use the `data` folder in the current path:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f173de8c-4255-442a-85cb-c46dbd114081",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-24 10:46:47,721 - INFO - Verified 'MedNIST.tar.gz', md5: 0bc7306e7427e00ad1c5526a6677552d.\n",
      "2024-05-24 10:46:47,722 - INFO - File exists: data/MedNIST.tar.gz, skipped downloading.\n",
      "2024-05-24 10:46:47,725 - INFO - Non-empty folder exists in data/MedNIST, skipped extracting.\n"
     ]
    }
   ],
   "source": [
    "root_dir = './'\n",
    "data_dir = './data/'\n",
    "os.makedirs(data_dir, exist_ok=True)\n",
    "\n",
    "train_data = MedNISTDataset(root_dir=data_dir, \n",
    "                            section=\"training\", \n",
    "                            download=True, \n",
    "                            progress=False, \n",
    "                            seed=0)\n",
    "train_datalist = [{\"image\": item[\"image\"]} for item in train_data.data if item[\"class_name\"] == \"Hand\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7224caae-3d4c-4161-907f-ad4119226233",
   "metadata": {},
   "source": [
    "Here we use transforms to augment the training dataset:\n",
    "\n",
    "- `LoadImaged` loads the hands images from files.\n",
    "- `EnsureChannelFirstd` ensures the original data to construct \"channel first\" shape.\n",
    "- `ScaleIntensityRanged` extracts intensity range [0, 255] and scales to [0, 1].\n",
    "- `RandAffined` efficiently performs rotate, scale, shear, translate, etc. together based on PyTorch affine transform."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "97d677ec-f294-4250-9c7e-fb069e74c38f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Loading dataset: 100%|████████████████████████████████████████████████████████████████████████████████████████| 7999/7999 [00:16<00:00, 476.45it/s]\n"
     ]
    }
   ],
   "source": [
    "train_transforms = transforms.Compose(\n",
    "    [\n",
    "        transforms.LoadImaged(keys=[\"image\"]),\n",
    "        transforms.EnsureChannelFirstd(keys=[\"image\"]),\n",
    "        transforms.ScaleIntensityRanged(keys=[\"image\"], \n",
    "                                        a_min=0.0, \n",
    "                                        a_max=255.0, \n",
    "                                        b_min=0.0, \n",
    "                                        b_max=1.0, \n",
    "                                        clip=True),\n",
    "        transforms.RandAffined(\n",
    "            keys=[\"image\"],\n",
    "            rotate_range=[(-np.pi / 36, np.pi / 36), (-np.pi / 36, np.pi / 36)],\n",
    "            translate_range=[(-1, 1), (-1, 1)],\n",
    "            scale_range=[(-0.05, 0.05), (-0.05, 0.05)],\n",
    "            spatial_size=[64, 64],\n",
    "            padding_mode=\"zeros\",\n",
    "            prob=0.5,\n",
    "        ),\n",
    "    ]\n",
    ")\n",
    "train_ds = CacheDataset(data=train_datalist, transform=train_transforms)\n",
    "train_loader = DataLoader(train_ds, batch_size=128, shuffle=True, num_workers=4, persistent_workers=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e5f0865-f4b8-42f6-9b45-17253e304a68",
   "metadata": {},
   "source": [
    "We can get the validation data the same way:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "98b2c37a-4c8f-4ea5-920b-8cf32ba9fba9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-24 10:49:18,525 - INFO - Verified 'MedNIST.tar.gz', md5: 0bc7306e7427e00ad1c5526a6677552d.\n",
      "2024-05-24 10:49:18,527 - INFO - File exists: data/MedNIST.tar.gz, skipped downloading.\n",
      "2024-05-24 10:49:18,528 - INFO - Non-empty folder exists in data/MedNIST, skipped extracting.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Loading dataset: 100%|████████████████████████████████████████████████████████████████████████████████████████| 1005/1005 [00:01<00:00, 670.05it/s]\n"
     ]
    }
   ],
   "source": [
    "val_data = MedNISTDataset(root_dir=data_dir, \n",
    "                          section=\"validation\", \n",
    "                          download=True, \n",
    "                          progress=False, \n",
    "                          seed=0)\n",
    "val_datalist = [{\"image\": item[\"image\"]} for item in val_data.data if item[\"class_name\"] == \"Hand\"]\n",
    "val_transforms = transforms.Compose(\n",
    "    [\n",
    "        transforms.LoadImaged(keys=[\"image\"]),\n",
    "        transforms.EnsureChannelFirstd(keys=[\"image\"]),\n",
    "        transforms.ScaleIntensityRanged(keys=[\"image\"], \n",
    "                                        a_min=0.0, \n",
    "                                        a_max=255.0, \n",
    "                                        b_min=0.0, \n",
    "                                        b_max=1.0, \n",
    "                                        clip=True),\n",
    "    ]\n",
    ")\n",
    "val_ds = CacheDataset(data=val_datalist, transform=val_transforms)\n",
    "val_loader = DataLoader(val_ds, batch_size=128, shuffle=False, num_workers=4, persistent_workers=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b24970a-6c8f-4177-a02a-053447fb9b70",
   "metadata": {},
   "source": [
    "### Visualisation of the training images\n",
    "\n",
    "Let see what the training images look like:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7684b200-bf11-4d8d-93c7-aa56b4350167",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "batch shape: torch.Size([128, 1, 64, 64])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "check_data = first(train_loader)\n",
    "print(f\"batch shape: {check_data['image'].shape}\")\n",
    "image_visualisation = torch.cat(\n",
    "    [check_data[\"image\"][0, 0], \n",
    "     check_data[\"image\"][1, 0], \n",
    "     check_data[\"image\"][2, 0], \n",
    "     check_data[\"image\"][3, 0]], dim=1\n",
    ")\n",
    "plt.figure(\"training images\", (12, 6))\n",
    "plt.imshow(image_visualisation, vmin=0, vmax=1, cmap=\"gray\")\n",
    "plt.axis(\"off\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20eb75e2-78e6-4117-afbc-2d670524e21b",
   "metadata": {},
   "source": [
    "## Define network, scheduler, optimizer, and inferer\n",
    "\n",
    "At this step, we instantiate the MONAI components to create a *DDPM*, the *UNET*, the *noise scheduler*, and the *inferer* used for training and sampling. \n",
    "\n",
    "We are using the original DDPM scheduler containing `1000` timesteps in its Markov chain, and a 2D UNET with attention mechanisms in the 2nd and 3rd levels, each with 1 attention head.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2deffe22-da47-44d1-bd18-beb23a8b6dea",
   "metadata": {},
   "outputs": [],
   "source": [
    "device = torch.device(\"cuda\")\n",
    "\n",
    "model = DiffusionModelUNet(\n",
    "    spatial_dims=2,\n",
    "    in_channels=1,\n",
    "    out_channels=1,\n",
    "    num_channels=(128, 256, 256),\n",
    "    attention_levels=(False, True, True),\n",
    "    num_res_blocks=1,\n",
    "    num_head_channels=256,\n",
    ")\n",
    "model.to(device)\n",
    "\n",
    "scheduler = DDPMScheduler(\n",
    "    num_train_timesteps=1000,\n",
    "    schedule=\"linear_beta\",\n",
    "    beta_start=0.0005,\n",
    "    beta_end=0.0195)\n",
    "\n",
    "optimizer = torch.optim.Adam(params=model.parameters(), lr=2.5e-5)\n",
    "\n",
    "inferer = DiffusionInferer(scheduler)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cca62fc2-a8ac-4ca3-828e-b02089a09ff0",
   "metadata": {},
   "source": [
    "## Model training\n",
    "\n",
    "Here, we are training our model for 75 epochs (training time: ~50 minutes).\n",
    "\n",
    "For tutorial, if you would like to skip the training and use a pre-trained model instead, set use_pretrained=True. This model was pretrained using the model in `pretrained/DDPM.pt`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "60971d87-bdaa-4418-bd19-ff0e143ac87e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Download start!\n",
      "Download Complete.\n"
     ]
    }
   ],
   "source": [
    "use_pretrained = True\n",
    "\n",
    "if use_pretrained:\n",
    "    os.makedirs('./pretrained/', exist_ok=True)\n",
    "    import urllib.request\n",
    "    url = 'https://raw.githubusercontent.com/vios-s/ISBI-2024-DiMEDIA-MONAI-Tutorial/main/pretrained/DDPM.pt'\n",
    "    print (\"Download start!\")\n",
    "    filename, headers = urllib.request.urlretrieve(url, filename=\"./pretrained/DDPM.pt\")\n",
    "    print(\"Download Complete.\")\n",
    "    model.load_state_dict(torch.load('pretrained/DDPM.pt'))\n",
    "    #model = torch.hub.load(\"marksgraham/pretrained_generative_models:v0.2\", model=\"ddpm_2d\", verbose=True).to(device)\n",
    "else:\n",
    "    n_epochs = 75\n",
    "    val_interval = 5\n",
    "    epoch_loss_list = []\n",
    "    val_epoch_loss_list = []\n",
    "\n",
    "    scaler = GradScaler()\n",
    "    total_start = time.time()\n",
    "    for epoch in range(n_epochs):\n",
    "        model.train()\n",
    "        epoch_loss = 0\n",
    "        progress_bar = tqdm(enumerate(train_loader), total=len(train_loader), ncols=70)\n",
    "        progress_bar.set_description(f\"Epoch {epoch}\")\n",
    "        for step, batch in progress_bar:\n",
    "            images = batch[\"image\"].to(device)\n",
    "            optimizer.zero_grad(set_to_none=True)\n",
    "\n",
    "            with autocast(enabled=True):\n",
    "                # Generate random noise\n",
    "                noise = torch.randn_like(images).to(device)\n",
    "\n",
    "                # Create timesteps\n",
    "                timesteps = torch.randint(\n",
    "                    0, \n",
    "                    inferer.scheduler.num_train_timesteps, \n",
    "                    (images.shape[0],), \n",
    "                    device=images.device\n",
    "                ).long()\n",
    "\n",
    "                # Get model prediction\n",
    "                noise_pred = inferer(\n",
    "                    inputs=images, \n",
    "                    diffusion_model=model, \n",
    "                    noise=noise, \n",
    "                    timesteps=timesteps)\n",
    "\n",
    "                loss = F.mse_loss(noise_pred.float(), noise.float())\n",
    "\n",
    "            scaler.scale(loss).backward()\n",
    "            scaler.step(optimizer)\n",
    "            scaler.update()\n",
    "\n",
    "            epoch_loss += loss.item()\n",
    "\n",
    "            progress_bar.set_postfix({\"loss\": epoch_loss / (step + 1)})\n",
    "        epoch_loss_list.append(epoch_loss / (step + 1))\n",
    "\n",
    "        if (epoch + 1) % val_interval == 0:\n",
    "            model.eval()\n",
    "            val_epoch_loss = 0\n",
    "            for step, batch in enumerate(val_loader):\n",
    "                images = batch[\"image\"].to(device)\n",
    "                with torch.no_grad():\n",
    "                    with autocast(enabled=True):\n",
    "                        noise = torch.randn_like(images).to(device)\n",
    "                        timesteps = torch.randint(\n",
    "                            0, inferer.scheduler.num_train_timesteps, (images.shape[0],), device=images.device\n",
    "                        ).long()\n",
    "                        noise_pred = inferer(inputs=images, diffusion_model=model, noise=noise, timesteps=timesteps)\n",
    "                        val_loss = F.mse_loss(noise_pred.float(), noise.float())\n",
    "\n",
    "                val_epoch_loss += val_loss.item()\n",
    "                progress_bar.set_postfix({\"val_loss\": val_epoch_loss / (step + 1)})\n",
    "            val_epoch_loss_list.append(val_epoch_loss / (step + 1))\n",
    "\n",
    "            # Sampling image during training\n",
    "            noise = torch.randn((1, 1, 64, 64))\n",
    "            noise = noise.to(device)\n",
    "            scheduler.set_timesteps(num_inference_steps=1000)\n",
    "            with autocast(enabled=True):\n",
    "                image = inferer.sample(input_noise=noise, diffusion_model=model, scheduler=scheduler)\n",
    "\n",
    "            plt.figure(figsize=(2, 2))\n",
    "            plt.imshow(image[0, 0].cpu(), vmin=0, vmax=1, cmap=\"gray\")\n",
    "            plt.tight_layout()\n",
    "            plt.axis(\"off\")\n",
    "            plt.show()\n",
    "\n",
    "    total_time = time.time() - total_start\n",
    "    print(f\"train completed, total time: {total_time}.\")\n",
    "    torch.save(model.state_dict(), 'pretrained/DDPM.pt')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84d55cb6-4957-487a-bc82-8e6e1d2e263e",
   "metadata": {},
   "source": [
    "## Plot the Learning Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3bede253-aa8b-4014-8924-cc7de9910ed7",
   "metadata": {},
   "outputs": [],
   "source": [
    "if not use_pretrained:\n",
    "    plt.style.use(\"seaborn-v0_8-white\")\n",
    "    plt.title(\"Learning Curves\", fontsize=20)\n",
    "    plt.plot(np.linspace(1, n_epochs, n_epochs), epoch_loss_list, color=\"C0\", linewidth=2.0, label=\"Train\")\n",
    "    plt.plot(\n",
    "        np.linspace(val_interval, n_epochs, int(n_epochs / val_interval)),\n",
    "        val_epoch_loss_list,\n",
    "        color=\"C1\",\n",
    "        linewidth=2.0,\n",
    "        label=\"Validation\",\n",
    "    )\n",
    "    plt.yticks(fontsize=12)\n",
    "    plt.xticks(fontsize=12)\n",
    "    plt.xlabel(\"Epochs\", fontsize=16)\n",
    "    plt.ylabel(\"Loss\", fontsize=16)\n",
    "    plt.legend(prop={\"size\": 14})\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7ab26f0-d0c3-478b-bea1-b2632933cd6a",
   "metadata": {},
   "source": [
    "## Plotting sampling process along DDPM's Markov chain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fbd50c41-4f18-4e11-bd3f-3c374e1fb24b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:12<00:00, 82.39it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.eval()\n",
    "noise = torch.randn((1, 1, 64, 64))\n",
    "noise = noise.to(device)\n",
    "scheduler.set_timesteps(num_inference_steps=1000)\n",
    "with autocast(enabled=True):\n",
    "    image, intermediates = inferer.sample(\n",
    "        input_noise=noise, diffusion_model=model, scheduler=scheduler, save_intermediates=True, intermediate_steps=100\n",
    "    )\n",
    "\n",
    "chain = torch.cat(intermediates, dim=-1)\n",
    "\n",
    "plt.style.use(\"default\")\n",
    "plt.imshow(chain[0, 0].cpu(), vmin=0, vmax=1, cmap=\"gray\")\n",
    "plt.tight_layout()\n",
    "plt.axis(\"off\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}