{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "ffc452d9-ab00-48e2-9e6a-038f0f57630a",
   "metadata": {},
   "source": [
    "# Create RO-Crate in ROHub for Wetland ML Demo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "25b2ee1f-beee-475a-8e38-f36d40729bf2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import rohub\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00d3e548-8cec-4ee7-86b8-a031e9511a40",
   "metadata": {},
   "source": [
    "## 1. AUTHENTICATION\n",
    "\n",
    "### Get your token from rohub.org → Settings → API Token"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "96aa7ebf-19d6-4e51-9cc6-7fc0e1ab7066",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Logged successfully as annef@simula.no.\n"
     ]
    }
   ],
   "source": [
    "rohub_user = open(os.path.join(os.environ['HOME'],\"rohub-user\")).read().rstrip()\n",
    "rohub_pwd = open(os.path.join(os.environ['HOME'],\"rohub-pwd\")).read().rstrip()\n",
    "rohub.login(username=rohub_user, password=rohub_pwd)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca5668a0-10ea-46d8-a50d-cc646472b7e5",
   "metadata": {},
   "source": [
    "## 2. CREATE RESEARCH OBJECT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0090be45-c64c-4d03-9a01-b9e05c5d05ee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Research Object was successfully created with id = 10dc322d-eedd-43ff-a4af-7adb6281cb6e\n",
      "Created RO: 10dc322d-eedd-43ff-a4af-7adb6281cb6e\n"
     ]
    }
   ],
   "source": [
    "ro = rohub.ros_create(\n",
    "    title=\"JupyterGIS Wetland ML Classification Demo - ESA EOPF\",\n",
    "    description=\"\"\"\n",
    "    Human-in-the-loop machine learning workflow for wetland classification \n",
    "    using Sentinel-2 data from ESA EOPF. Demonstrates collaborative annotation \n",
    "    using JupyterGIS, model retraining with expert corrections, and FAIR \n",
    "    research practices.\n",
    "    \"\"\",\n",
    "    ros_type=\"Executable Research Object\",\n",
    "    use_template=True,\n",
    "    research_areas=[\"Earth Observation\", \"Machine Learning\", \"Wetlands\", \"Remote Sensing\"]\n",
    ")\n",
    "\n",
    "print(f\"Created RO: {ro.identifier}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d84ada9a-5e6d-4ae1-89fc-b7e2014394f2",
   "metadata": {},
   "source": [
    "## 3. ADD EXTERNAL RESOURCES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1c69795f-00a1-485e-be5b-8beb577dd319",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Resource was successfully created with id = 6120644a-edf9-4139-ad55-96df61c61785\n",
      "Resource was successfully created with id = 98b35d80-4f51-45c1-a3bf-eeb8685337e5\n",
      "Resource was successfully created with id = 12b9f9ef-a046-4222-aea1-8e806706a3f1\n",
      "Resource was successfully created with id = 2f956d0f-5f9f-4768-8fe6-f6dd249ac856\n",
      "Resource was successfully created with id = c267902c-8c44-464d-b224-f72fb04a0a30\n",
      "Resource was successfully created with id = 4d4a1bbc-2c18-44d5-b3c1-df9a22c5e1cd\n",
      "Resource was successfully created with id = a28552fd-7d9c-449b-98a4-884da477ae2b\n",
      "Resource was successfully created with id = acd096ac-59d0-4925-bb2a-3c3c50c2fc31\n",
      "Resource was successfully created with id = 591a9965-170b-4ef9-ac42-5f0aa4b26fdd\n",
      "Resource was successfully created with id = 11e0f30e-32bc-4aca-9c45-20244e641b68\n",
      "Resource was successfully created with id = 3e7c2e75-e59b-466a-ba56-600ff9cace09\n",
      "All resources added!\n"
     ]
    }
   ],
   "source": [
    "GITHUB_RAW = \"https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content\"\n",
    "GITHUB_HTML = \"https://annefou.github.io/jupytergis-showcases/lab/index.html\"\n",
    "\n",
    "# Software / Tools\n",
    "jupytergis = ro.add_external_resource(\n",
    "    res_type=\"Software\",\n",
    "    input_url=\"https://github.com/geojupyter/jupytergis\",\n",
    "    folder=\"tool\",\n",
    "    title=\"JupyterGIS\",\n",
    "    description=\"Collaborative GIS environment for Jupyter - required to open .jGIS files\"\n",
    ")\n",
    "\n",
    "conda_env = ro.add_external_resource(\n",
    "    res_type=\"Software\",\n",
    "    input_url=f\"{GITHUB_RAW}/../requirements.txt\",\n",
    "    folder=\"tool\",\n",
    "    title=\"Conda Environment\",\n",
    "    description=\"Conda environment specification with all Python dependencies\"\n",
    ")\n",
    "\n",
    "eopf_service = ro.add_external_resource(\n",
    "    res_type=\"Other\",\n",
    "    input_url=\"https://github.com/EOPF-Sample-Service/eopf-sample-notebooks\",\n",
    "    folder=\"biblio\",\n",
    "    title=\"EOPF Sample Service\",\n",
    "    description=\"ESA Earth Observation Processing Framework for Sentinel-1, 2 and 3 data access\"\n",
    ")\n",
    "\n",
    "# Workflow (main entry point)\n",
    "notebook = ro.add_external_resource(\n",
    "    res_type=\"Jupyter Notebook\",\n",
    "    input_url=f\"{GITHUB_RAW}/Wetland_ML_Demo_EOPF.ipynb\",\n",
    "    folder=\"tool\",\n",
    "    title=\"Wetland ML Demo Notebook\",\n",
    "    description=\"Main Jupyter notebook implementing the wetland classification workflow\"\n",
    ")\n",
    "\n",
    "# JupyterGIS document\n",
    "jgis_doc = ro.add_external_resource(\n",
    "    res_type=\"Dataset\",\n",
    "    input_url=f\"{GITHUB_HTML}?path=Wetland_Annotation.jGIS\",\n",
    "    folder=\"output\",\n",
    "    title=\"JupyterGIS Annotation Document\",\n",
    "    description=\"Interactive map with expert annotations for model corrections\"\n",
    ")\n",
    "\n",
    "# Input data\n",
    "sentinel2_rgb = ro.add_external_resource(\n",
    "    res_type=\"Dataset\",\n",
    "    input_url=f\"{GITHUB_RAW}/wetland_outputs/sentinel2_rgb.tif\",\n",
    "    folder=\"output\",\n",
    "    title=\"Sentinel-2 RGB Composite (COG)\",\n",
    "    description=\"Cloud Optimized GeoTIFF - RGB composite from Sentinel-2 L2A\"\n",
    ")\n",
    "\n",
    "# ML outputs - Version 1\n",
    "prediction_v1 = ro.add_external_resource(\n",
    "    res_type=\"Result\",\n",
    "    input_url=f\"{GITHUB_RAW}/wetland_outputs/wetland_prediction_v1.tif\",\n",
    "    folder=\"output\",\n",
    "    title=\"Wetland Prediction v1\",\n",
    "    description=\"Initial Random Forest classification - before expert corrections\"\n",
    ")\n",
    "\n",
    "# Corrections\n",
    "corrections = ro.add_external_resource(\n",
    "    res_type=\"Dataset\",\n",
    "    input_url=f\"{GITHUB_RAW}/wetland_outputs/corrections.geojson\",\n",
    "    folder=\"input\",\n",
    "    title=\"Expert Corrections (GeoJSON)\",\n",
    "    description=\"Expert corrections extracted from JupyterGIS annotations\"\n",
    ")\n",
    "\n",
    "# ML outputs - Version 2\n",
    "prediction_v2 = ro.add_external_resource(\n",
    "    res_type=\"Result\",\n",
    "    input_url=f\"{GITHUB_RAW}/wetland_outputs/wetland_prediction_v2_corrected.tif\",\n",
    "    folder=\"output\",\n",
    "    title=\"Wetland Prediction v2 (Corrected)\",\n",
    "    description=\"Improved classification after retraining with expert corrections\"\n",
    ")\n",
    "\n",
    "model_v2 = ro.add_external_resource(\n",
    "    res_type=\"Result\",\n",
    "    input_url=f\"{GITHUB_RAW}/wetland_outputs/wetland_model_v2.joblib\",\n",
    "    folder=\"output\",\n",
    "    title=\"Trained Model v2 (joblib)\",\n",
    "    description=\"Serialized Random Forest model retrained with expert corrections\"\n",
    ")\n",
    "\n",
    "# GitHub repository\n",
    "github_repo = ro.add_external_resource(\n",
    "    res_type=\"Software\",\n",
    "    input_url=\"https://github.com/annefou/jupytergis-showcases\",\n",
    "    folder=\"tool\",\n",
    "    title=\"GitHub Repository\",\n",
    "    description=\"Source repository for this demo\"\n",
    ")\n",
    "\n",
    "print(\"All resources added!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d991ac54-8540-4413-adbf-1a098a922d4c",
   "metadata": {},
   "source": [
    "## 4. CREATE LOOKUP FOR PROVENANCE TRIPLES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1a65cdbd-550b-4583-ae65-b1cc78cb3e2a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Total resources: 11\n",
      "   - JupyterGIS Annotation Document\n",
      "   - GitHub Repository\n",
      "   - EOPF Sample Service\n",
      "   - JupyterGIS\n",
      "   - Conda Environment\n",
      "   - Wetland ML Demo Notebook\n",
      "   - Expert Corrections (GeoJSON)\n",
      "   - Sentinel-2 RGB Composite (COG)\n",
      "   - Trained Model v2 (joblib)\n",
      "   - Wetland Prediction v1\n",
      "   - Wetland Prediction v2 (Corrected)\n"
     ]
    }
   ],
   "source": [
    "# Get resources with their URLs\n",
    "resources = ro.list_resources()\n",
    "res_url_by_title = dict(zip(resources['title'], resources['url']))\n",
    "\n",
    "print(f\"\\nTotal resources: {len(res_url_by_title)}\")\n",
    "for title in res_url_by_title:\n",
    "    print(f\"   - {title}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "724c3c38-de94-408a-a9e4-7798cf3928bf",
   "metadata": {},
   "source": [
    "## 5. ADD PROVENANCE RELATIONSHIPS (TRIPLES)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "689ab4ac-61a5-4e4d-aa84-ed2a3c3c66c2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Added: Notebook -> softwareRequirements -> Conda Environment\n",
      "Added: Corrections -> wasDerivedFrom -> jGIS Document\n",
      "Added: Prediction v2 -> wasDerivedFrom -> Corrections\n",
      "Added: Prediction v2 -> wasRevisionOf -> Prediction v1\n",
      "Added: Prediction v1 -> wasGeneratedBy -> Notebook\n"
     ]
    }
   ],
   "source": [
    "# Use the first annotation to attach triples\n",
    "annotations = ro.list_annotations()\n",
    "annotation_id = annotations[0]['identifier']\n",
    "\n",
    "# 1. Notebook requires Conda Environment\n",
    "ro.add_triple(\n",
    "    the_subject=res_url_by_title[\"Wetland ML Demo Notebook\"],\n",
    "    the_predicate=\"http://schema.org/softwareRequirements\",\n",
    "    the_object=res_url_by_title[\"Conda Environment\"],\n",
    "    annotation_id=annotation_id,\n",
    "    object_class=\"URIRef\"\n",
    ")\n",
    "print(\"Added: Notebook -> softwareRequirements -> Conda Environment\")\n",
    "\n",
    "# 2. Corrections derived from JupyterGIS annotations\n",
    "ro.add_triple(\n",
    "    the_subject=res_url_by_title[\"Expert Corrections (GeoJSON)\"],\n",
    "    the_predicate=\"http://www.w3.org/ns/prov#wasDerivedFrom\",\n",
    "    the_object=res_url_by_title[\"JupyterGIS Annotation Document\"],\n",
    "    annotation_id=annotation_id,\n",
    "    object_class=\"URIRef\"\n",
    ")\n",
    "print(\"Added: Corrections -> wasDerivedFrom -> jGIS Document\")\n",
    "\n",
    "# 3. Prediction v2 derived from corrections\n",
    "ro.add_triple(\n",
    "    the_subject=res_url_by_title[\"Wetland Prediction v2 (Corrected)\"],\n",
    "    the_predicate=\"http://www.w3.org/ns/prov#wasDerivedFrom\",\n",
    "    the_object=res_url_by_title[\"Expert Corrections (GeoJSON)\"],\n",
    "    annotation_id=annotation_id,\n",
    "    object_class=\"URIRef\"\n",
    ")\n",
    "print(\"Added: Prediction v2 -> wasDerivedFrom -> Corrections\")\n",
    "\n",
    "# 4. Prediction v2 is revision of v1\n",
    "ro.add_triple(\n",
    "    the_subject=res_url_by_title[\"Wetland Prediction v2 (Corrected)\"],\n",
    "    the_predicate=\"http://www.w3.org/ns/prov#wasRevisionOf\",\n",
    "    the_object=res_url_by_title[\"Wetland Prediction v1\"],\n",
    "    annotation_id=annotation_id,\n",
    "    object_class=\"URIRef\"\n",
    ")\n",
    "print(\"Added: Prediction v2 -> wasRevisionOf -> Prediction v1\")\n",
    "\n",
    "# 5. Prediction v1 generated by notebook\n",
    "ro.add_triple(\n",
    "    the_subject=res_url_by_title[\"Wetland Prediction v1\"],\n",
    "    the_predicate=\"http://www.w3.org/ns/prov#wasGeneratedBy\",\n",
    "    the_object=res_url_by_title[\"Wetland ML Demo Notebook\"],\n",
    "    annotation_id=annotation_id,\n",
    "    object_class=\"URIRef\"\n",
    ")\n",
    "print(\"Added: Prediction v1 -> wasGeneratedBy -> Notebook\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "49ac70d0-db02-4278-b3af-04f544caca94",
   "metadata": {},
   "source": [
    "## 6. VERIFY TRIPLES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "8f0b956d-bc79-4025-963c-5c2076e457d8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'identifier': '09dad437-4cb5-4cfe-88e2-72046bc8247a', 'subject': 'https://w3id.org/ro-id/10dc322d-eedd-43ff-a4af-7adb6281cb6e/resources/3e7c2e75-e59b-466a-ba56-600ff9cace09', 'predicate': 'http://purl.org/dc/terms/title', 'object': 'GitHub Repository', 'created_by': 'annef@simula.no', 'created_on': '2025-12-21T16:29:33.509179+0000'}, {'identifier': '0589188d-a60d-42ea-9402-065903e1d142', 'subject': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/Wetland_ML_Demo_EOPF.ipynb', 'predicate': 'http://schema.org/softwareRequirements', 'object': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/../requirements.txt', 'created_by': 'annef@simula.no', 'created_on': '2025-12-21T16:29:50.643195+0000'}, {'identifier': 'e9b8478f-d77b-46ec-a801-7c53ae4bac1b', 'subject': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/wetland_outputs/corrections.geojson', 'predicate': 'http://www.w3.org/ns/prov#wasDerivedFrom', 'object': 'https://annefou.github.io/jupytergis-showcases/lab/index.html?path=Wetland_Annotation.jGIS', 'created_by': 'annef@simula.no', 'created_on': '2025-12-21T16:29:51.040155+0000'}, {'identifier': '1557ef0c-3183-4a44-951d-ae2e7dac46fb', 'subject': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/wetland_outputs/wetland_prediction_v2_corrected.tif', 'predicate': 'http://www.w3.org/ns/prov#wasDerivedFrom', 'object': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/wetland_outputs/corrections.geojson', 'created_by': 'annef@simula.no', 'created_on': '2025-12-21T16:29:51.433140+0000'}, {'identifier': '4114351c-e325-4363-8cae-7ab351f3b8e5', 'subject': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/wetland_outputs/wetland_prediction_v1.tif', 'predicate': 'http://www.w3.org/ns/prov#wasGeneratedBy', 'object': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/Wetland_ML_Demo_EOPF.ipynb', 'created_by': 'annef@simula.no', 'created_on': '2025-12-21T16:29:52.233198+0000'}, {'identifier': 'ebcd3f50-7154-4f5e-a7b1-b3f1e4227f1d', 'subject': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/wetland_outputs/wetland_prediction_v2_corrected.tif', 'predicate': 'http://www.w3.org/ns/prov#wasRevisionOf', 'object': 'https://raw.githubusercontent.com/annefou/jupytergis-showcases/refs/heads/main/content/wetland_outputs/wetland_prediction_v1.tif', 'created_by': 'annef@simula.no', 'created_on': '2025-12-21T16:29:51.836535+0000'}]\n"
     ]
    }
   ],
   "source": [
    "# Check triples in the annotation\n",
    "triples = ro.list_triples(annotation_id)\n",
    "print(triples)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a238e7d4-1683-4816-b0b6-d0e4e21aba33",
   "metadata": {},
   "source": [
    "## 7. ADD AUTHOR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0323e2b8-44a8-45b1-b7cb-88d3acf4b4a0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Agent: annef@simula.no recognized in the system.\n",
      "Author added!\n"
     ]
    }
   ],
   "source": [
    "ro.set_authors(agents=[\"annef@simula.no\"])\n",
    "print(\"Author added!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91efd76d-ee00-4458-9213-8b05bc2f40af",
   "metadata": {},
   "source": [
    "## 8. EXPORT RO-CRATE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "250ffc95-f67b-444e-9b67-ee277ff94752",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "File was successfully downloaded.\n"
     ]
    }
   ],
   "source": [
    "# Export the RO-Crate to verify the relationships are captured\n",
    "ro.export_to_rocrate(filename=\"my_rocrate\", use_format=\"jsonld\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "add9ecbf-8b7d-4aa6-8a28-8c1148517709",
   "metadata": {},
   "source": [
    "## 8. SUMMARY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5f22f458-42a1-430b-a383-85667748d73c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=================================================================\n",
      "RESEARCH OBJECT CREATED SUCCESSFULLY\n",
      "=================================================================\n",
      "\n",
      "📊 Provenance graph:\n",
      "\n",
      "   EOPF Service ──────────────────┐\n",
      "         │                        │\n",
      "         ▼                        ▼\n",
      "   Sentinel-2 RGB          Notebook (Workflow)\n",
      "                                  │\n",
      "                                  ▼\n",
      "                           Prediction v1\n",
      "                                  │\n",
      "         ┌────────────────────────┤\n",
      "         ▼                        ▼\n",
      "   JupyterGIS ◀─────────── jGIS Document\n",
      "   (Software)                     │\n",
      "                                  ▼\n",
      "                           Corrections (GeoJSON)\n",
      "                                  │\n",
      "         ┌────────────────────────┤\n",
      "         ▼                        ▼\n",
      "   Model v2 ──────────────▶ Prediction v2\n",
      "\n",
      "🔗 View your RO at:\n",
      "   https://w3id.org/ro-id/10dc322d-eedd-43ff-a4af-7adb6281cb6e\n",
      "   https://rohub.org/overview/10dc322d-eedd-43ff-a4af-7adb6281cb6e\n",
      "\n",
      "=================================================================\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f\"\"\"\n",
    "{'='*65}\n",
    "RESEARCH OBJECT CREATED SUCCESSFULLY\n",
    "{'='*65}\n",
    "\n",
    "📊 Provenance graph:\n",
    "\n",
    "   EOPF Service ──────────────────┐\n",
    "         │                        │\n",
    "         ▼                        ▼\n",
    "   Sentinel-2 RGB          Notebook (Workflow)\n",
    "                                  │\n",
    "                                  ▼\n",
    "                           Prediction v1\n",
    "                                  │\n",
    "         ┌────────────────────────┤\n",
    "         ▼                        ▼\n",
    "   JupyterGIS ◀─────────── jGIS Document\n",
    "   (Software)                     │\n",
    "                                  ▼\n",
    "                           Corrections (GeoJSON)\n",
    "                                  │\n",
    "         ┌────────────────────────┤\n",
    "         ▼                        ▼\n",
    "   Model v2 ──────────────▶ Prediction v2\n",
    "\n",
    "🔗 View your RO at:\n",
    "   https://w3id.org/ro-id/{ro.identifier}\n",
    "   https://rohub.org/overview/{ro.identifier}\n",
    "\n",
    "{'='*65}\n",
    "\"\"\")"
   ]
  },
  {
   "cell_type": "raw",
   "id": "d31b4e23-9d28-4c01-9c6a-fb6918380f61",
   "metadata": {},
   "source": [
    "ro_id = \"e62f59cd-fe68-4ca5-9656-9411d82f8b28\"\n",
    "#rohub.ros_delete(identifier = ro_id)\n",
    "\n",
    "ro = rohub.ros_load(identifier=ro_id)\n",
    "print(ro.status)\n",
    "print(ro.creation_mode)\n",
    "ro.delete()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a0c37150-786b-48e8-b0c0-691b6563efd3",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bc37d2ad-b740-408a-97ee-691e807d3a82",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "900fa66d-78f3-4379-a345-3ef400c01bbf",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.12.12"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}