Pyronear Risks
The pyro-risks project aims at providing the pyronear-platform with a machine learning based wildfire forecasting capability.
Table of Contents
Getting started
Prerequisites
- Python 3.6 (or more recent)
- pip
Installation
You can install the package from github as follows:
pip install git+https://github.com/pyronear/pyro-risks
Usage
Beforehand, you will need to set a few environment variables either manually or by writing an .env
file in the root directory of this project, like in the example below:
CDS_UID=my_secret_uid
CDS_API_KEY=my_very_secret_key
Those values will allow your web server to connect to CDS API, which is mandatory for your datasets access to be fully operational.
Web server
To be able to expose model inference, you can run a web server using docker containers with this command:
PORT=8003 docker-compose up -d --build
Once completed, you will notice that you have a docker container running on the port you selected, which can process requests just like any web server.
Examples
datasets
Access the main pyro-risks datasets locally.
from pyro_risks.datasets import NASAFIRMS, NASAFIRMS_VIIRS, GwisFwi, ERA5T, ERALand
modis = NASAFIRMS()
viirs = NASAFIRMS_VIIRS()
fdi = GwisFwi()
era = ERA5T()
era_land = ERA5Land()
Scripts
You are free to merge the datasets however you want and to implement any zonal statistic you want, but some are already provided for reference. In order to use them check the example scripts options as follows:
python scripts/example_ERA5_FIRMS.py --help
You can then run the script with your own arguments:
python scripts/example_ERA5_FIRMS.py --type_of_merged departements
Documentation
The full package documentation is available here for detailed specifications. The documentation was built with Sphinx using a theme provided by Read the Docs.
Contributing
Please refer to the CONTRIBUTING
guide if you wish to contribute to this project.
Credits
This project is developed and maintained by the repo owner and volunteers from Data for Good.
License
Distributed under the Apache v2 License. See LICENSE
for more information.