Official repository for the SwAIther-Precip paper.
-
Updated
Jun 8, 2026 - Python
Official repository for the SwAIther-Precip paper.
Meteorological data from radar and satellite sources often contain noise due to air turbulence and device manipulation, leading to inaccuracies in predictions. Mitigating with ANN & LSTM
Machine learning pipeline for monthly precipitation forecasting in South America using atmospheric data and ensemble learning.
Neural-network architecture for probabilistic multi-source precipitation forecasting over the Netherlands
Previsão mensal de precipitação na América do Sul: LightGBM, memória atmosférica e validação temporal. Projeto WorCAP 2026 / INPE.
Python training pipeline for fitting an sklearn MLP regression model on sequential rainfall data and exporting the artifact alongside preprocessing scaler configurations to ONNX format.
This repository contains the source code, synthetic benchmark suites, and documentation associated with the manuscript: "MODE-py: A Python framework for spatio-temporal object-based verification of high resolution precipitation forecasts" (Submitted to Computers & Geosciences).
R code and reproducibility materials for probabilistic hourly precipitation forecasting using ZAGA GAMLSS boosting.
To associate your repository with the precipitation-forecasting topic, visit your repo's landing page and select "manage topics."