wass2s documentation

A python-based tool for seasonal climate forecast in West Africa and the Sahel.

The wass2s tool is designed to facilitate implementation of the new generation of seasonal forecasts in West Africa and the Sahel using various statistical and machine learning methods. New generation of seasonal forecasts aligns with the World Meteorological Organization’s (WMO) guidelines for objective, operational, and scientifically rigorous seasonal forecasting methods. wass2s helps forecaster to download GCM, reanalysis, and satellite/observation data, build statistical or machine learning models, verify the models using cross-validation, and forecast. A user-friendly jupyter-lab notebook streamlines the forecasting process .

Features

• Automated Forecasting

• Reproducibility

• Modularity

• Exploration of Machine Learning Models.

Contents:

• Installation
• Usage
  • Download module
  • Processing Modules
  • Quantifying uncertainty via cross-validation
  • Models Modules
  • Forecast
  • Verification Module
  • Multi-Model Ensemble (MME) Techniques
• wass2s submodules
  • wass2s.was_download module
  • wass2s.was_transformdata module
  • wass2s.was_compute_predictand module
  • wass2s.was_bias_correction module
  • wass2s.was_merge_predictand module
  • wass2s.was_cross_validate module
  • wass2s.was_linear_models module
  • wass2s.was_eof module
  • wass2s.was_pcr module
  • wass2s.was_cca module
  • wass2s.was_machine_learning module
  • wass2s.was_analog module
  • wass2s.was_verification module
  • wass2s.was_mme module
  • wass2s.utils module