The increasing adoption of mobile phones have significantly shortened the time required for firefighting agents to be alerted of a starting wildfire. In less dense areas, limiting and minimizing this duration remains critical to preserve forest areas.
PyroNear aims at providing the means to create a wildfire early detection system with state-of-the-art performances at minimal deployment costs.
- Python 3.6 (or more recent)
- pip
You can install the package using pypi as follows:
pip install pyronear
Access all PyroNear datasets just like any torchvision.datasets.VisionDataset
:
from pyronear.datasets import OpenFire
dataset = OpenFire('./data', download=True)
You are free to use any training script, but some are already provided for reference. In order to use them, install the specific requirements and check script options as follows:
pip install -r references/classification/fastai/requirements.txt
python references/classification/fastai/train.py --help
You can then run the script with your own arguments:
python references/classification/fastai/train.py --lr 3e-3 --epochs 4 --pretrained --deterministic
Please note that most tasks are provided with two training scripts (and their requirements.txt
): one using fastai and the other without it.
The full package documentation is available here for detailed specifications. The documentation was built with Sphinx using a theme provided by Read the Docs.
Please refer to CONTRIBUTING
if you wish to contribute to this project.
This project is developed and maintained by the repo owner and volunteers from Data for Good.
Distributed under the MIT License. See LICENSE
for more information.