Rain nowcast

Encadrants : Dominique Béréziat
Disponible : NON
Spécialité : IMA
Nombre d'étudiants : 2
Description :

Rainfall forecasting is an important environmental issue (flood anticipation, agricultural management, …). Here we are interested in a short-term forecast (time horizon up to 2 hours). We use Radar data which are robust rainfall indicators. The algorithmic will be based on Deep learning approaches with a focus on convolutional LSTM architectures.

Pré-requis : Python and machine learning tools
Travail demandé : - state of the art and critical analysis
- implementation in Pytorch or Keras on Meteo-France data