Jupyter Test: FastAPI + JupyterLab NDVI Notebook Runner

This project provides a FastAPI web service that executes a parameterized Jupyter notebook (for NDVI calculation) using Papermill and allows you to view the results in JupyterLab.

Features

• Run a notebook remotely: Trigger execution of a notebook with a custom GeoTIFF URL and optional bounding box (bbox).
• View results in JupyterLab: Redirects to a JupyterLab instance to view the output notebook.

Requirements

• Python 3.13+
• JupyterLab
• Uvicorn (for running FastAPI)
• All dependencies in pyproject.toml (install with your preferred tool, e.g. pip, uv, or poetry)

Installation

1. Clone the repository and navigate to the project directory.

2. Install dependencies (using pip as an example):

   pip install fastapi jupyterlab papermill rioxarray matplotlib uvicorn

   Or, to use the exact versions in pyproject.toml:

   pip install -r requirements.txt
   # or use uv/poetry as appropriate

Usage

1. Start JupyterLab

You must have a JupyterLab server running so the FastAPI app can redirect you to view the output notebooks.

jupyter lab --notebook-dir=notebooks --port=8889

• The --notebook-dir=notebooks flag ensures output notebooks are saved and accessible in the correct folder.
• The --port=8889 flag matches the redirect URL in main.py.

2. Start the FastAPI App

Run the FastAPI app using Uvicorn:

uvicorn main:app --reload

• By default, this will start the API at http://127.0.0.1:8000.

3. Run a Notebook

Open your browser and visit:

http://127.0.0.1:8000/run-notebook?cog_url=YOUR_COG_URL&bbox=min_lon,min_lat,max_lon,max_lat

• Replace YOUR_COG_URL with the URL to your GeoTIFF file.
• The bbox parameter is optional. If provided, it should be four comma-separated values: min_lon,min_lat,max_lon,max_lat (in WGS84/EPSG:4326 coordinates).
• Example:

http://127.0.0.1:8000/run-notebook?cog_url=https%3A%2F%2Fdap.ceda.ac.uk%2Fneodc%2Fsentinel_ard%2Fdata%2Fsentinel_2%2F2025%2F06%2F20%2FS2B_20250620_latn537lonw0037_T30UVE_ORB080_20250620132615_utm30n_osgb_vmsk_sharp_rad_srefdem_stdsref.tif&bbox=-4.531313295277459,53.151777988193025,-4.5213132952774595,53.16177798819302

• This will execute the templates/ndvi_calculation.ipynb notebook with the provided parameters.
• After execution, you will be redirected to JupyterLab to view the output notebook.

File Structure

• main.py — FastAPI app for running and viewing notebooks
• templates/ndvi_calculation.ipynb — Parameterized notebook template
• notebooks/ — Output notebooks are saved here
• pyproject.toml — Project dependencies

Notes

• Make sure the JupyterLab server is running and accessible at http://localhost:8889/lab/tree/notebooks/.
• The FastAPI app does not serve the notebook files directly; it only redirects to JupyterLab.
• The notebook will use the provided bbox if given, otherwise it will use a default bounding box.

Troubleshooting

• If you get a 404 when redirected, ensure JupyterLab is running and the output notebook exists in the notebooks/ directory.
• If you see errors about missing dependencies, double-check your Python environment and install all required packages.

---