{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# MONAI Label with nnUNet-MONAI Bundle\n",
    "\n",
    "This notebook demonstrates how to use the MONAI Label with nnUNet-MONAI Bundle to perform active learning and Auto-Segmentation with the nnUNet-MONAI Bundle.\n",
    "\n",
    "## Start MONAI Label\n",
    "\n",
    "To start MONAI Label with the nnUNet-MONAI Bundle App, run the following command:\n",
    "\n",
    "```bash\n",
    " monailabel start_server --app nnunetmonaibundle --studies /home/maia-user/Tutorials/MAIA/Task09_Spleen/imagesTr --conf models Task09_Spleen_Bundle\n",
    "```\n",
    "\n",
    "and access the MONAI Label App at http://localhost:8000\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train Task\n",
    "\n",
    "The Train Task requires the following parameters:\n",
    "\n",
    "- `max_epochs`: Total number of epochs to run\n",
    "- `tracking_uri`: MLFlow Tracking URI to monitor the training\n",
    "- `nnunet_plans_identifier`: nnUNet plans identifier, specific for the experiment (e.g. nnUNetPlans, nnUNetResEncUNetLPlans, nnUNetResEncUNetPlans)\n",
    "- `nnunet_planner_name`: nnUNet planner name, specific for the experiment (e.g. nnUNetPlanner, nnUNetPlannerResEnc, nnUNetPlannerResEncL)\n",
    "- `nnunet_root_folder`: Root folder where the nnUNet experiments are stored\n",
    "- `task_name`: nnUNet Task Name\n",
    "- `task_id`: nnUNet Task ID\n",
    "- `pymaia_config_file`: PyMaia configuration file (see [Prepare nnUNet ResEnc Training](MAIA.ipynb#Prepare-nnUNet-ResEnc-Training))\n",
    "\n",
    "Internally, the nnUNetBundle Train Task will run the following steps:\n",
    "\n",
    "1. [Prepare nnUNet Experiment](MAIA.ipynb#Prepare-nnUNet-Experiment)\n",
    "2. [nnUNet Plan and Preprocessing](MAIA.ipynb#nnUNet-Plan-and-Preprocessing)\n",
    "3. [nnUNet Training](MAIA.ipynb#nnUNet-Training)\n",
    "4. [Convert nnUNet MONAI Bundle to native nnUNet for Inference](MAIA.ipynb#Validation)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Inference Task\n",
    "\n",
    "\n",
    "The Auto-Segmentation action can be performed when a valid model is available in the MONAI Bundle `models` folder.\n",
    "\n",
    "A valid `model_folder` parameter should be specified in the `inference.yaml`file, referring to the nnUNet model folder directory (`<TRAINER_CLASS__PLANS_IDENTIFIER__3d_fullres>`), inside which the folds subdirectories with the model checkpoint are located:\n",
    "\n",
    "```plaintext\n",
    "model_folder/\n",
    "├── fold_0/\n",
    "│   ├── checkpoint_final.pth\n",
    "│   └── checkpoint_best.pth\n",
    "├── fold_1/\n",
    "│   ├── checkpoint_final.pth\n",
    "│   └── checkpoint_best.pth\n",
    "├── plans.json\n",
    "└── dataset.json\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load existing nnUNet Model into nnUNet-MONAI Bundle\n",
    "\n",
    "\n",
    "To load an existing nnUNet trained model into the nnUNet-MONAI Bundle:\n",
    "\n",
    "```bash\n",
    "nnUNetv2_install_pretrained_model_from_zip --zip /path/to/nnUNet_pretrained_model.zip\n",
    "```\n",
    "and modify the corresponding `model_folder` parameter in the `inference.yaml` file.\n",
    "\n",
    "By default, the pretrained model will be extracted under the `$nnUNet_results` directory."
   ]
  }
 ],
 "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}