Skip to content

Python library for radar-based precipitation nowcasting based on optical flow techniques

License

Notifications You must be signed in to change notification settings

hydrogo/rainymotion

Repository files navigation

rainymotion

rainymotion logo

rainymotion: Python library for radar-based precipitation nowcasting based on optical flow techniques

Idea

The main idea of the rainymotion library is to provide an open baseline solution for radar-based precipitation nowcasting.

Development

rainymotion is based only on free and open source software -- we tried to make a clue between the best scientific libraries to make them work together for providing reliable precipitation nowcasts.

rainymotion logo

Documentation

rainymotion documentation is hosted by Read the Docs.

Example

To obtain precipitation nowcasts using the rainymotion models you need to follow the only three steps:

  1. initialize the model (you have 4 variants)
  2. load data to the model's placeholder
  3. run the model
# import rainymotion model
from rainymotion.models import Dense

# initialize model instance
model = Dense()

# load the data using your custom DataLoader function
model.input_data = DataLoader("/path/to/data")

# run the model
nowcasts = model.run()

Reference

Please cite rainymotion as Ayzel, G., Heistermann, M., and Winterrath, T.: Optical flow models as an open benchmark for radar-based precipitation nowcasting (rainymotion v0.1), Geosci. Model Dev., 12, 1387-1402, https://doi.org/10.5194/gmd-12-1387-2019, 2019.

About

Python library for radar-based precipitation nowcasting based on optical flow techniques

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages