This repository contains API backend supporting icenet-dashboard
. It is built on top of Titiler and FastAPI to serve Cloud Optimized GeoTIFF (COG) files from icenet.
This application is designed to be used in conjunction with icenet-dashboard. It serves COG files and STAC catalogs from a configurable directory, which can be mounted as a volume or passed via an environment variable.
This codebase contains two apps which are to be run concurrently, app.py
and serve_data.py
.
-
TiTiler-based FastAPI application, including route definitions for COG and STAC tilers.
-
FastAPI server to serve static data from a configurable directory (by default,
./data
). This is used to serve the static files, such as STAC catalogs and COG files.
If you are attempting to run this independent of the orchestrating repository (icenet-dashboard-meta
), then, make sure to first convert the IceNet netCDF predictions to COG using icenet-dashboard-preprocessor
first which will create a data/
directory at the root of this repository (assuming you're running the conversion command from this path). Then, you can run this independent of the orchestrator repo.
git clone https://github.com/icenet-ai/icenet-tiler-api.git
cd icenet-tiler-api
pip install -r requirements.txt
make run
This will run both app.py
and serve_data.py
across ports 8000
and 8002
respectively. The API docs can be accessed at http://localhost:8000/docs.