Stars
RAFConv: Innovating Spatital Attention and Standard Convolutional Operation
TianQuan-Climate, a novel machine learning model that provides accurate global daily mean forecasts up to 45 days by integrating climatology mean state information.
Bridging KAN and MLP: MJKAN, a hybrid architecture with both efficiency and expressiveness
[CVPR26] TESSERA is a foundation model that can process time-series satellite imagery for applications such as land classification and canopy height prediction. Developed at the University of Cambr…
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
Software for training, running and analyzing a Deep Learning Earth System Model (DLESyM)
MambaOut: Do We Really Need Mamba for Vision? (CVPR 2025)
[NeurIPS'25 Spotlight] Official repository for "Chain-of-Zoom: Extreme Super-Resolution via Scale Autoregression and Preference Alignment"
[AAAI 2026] The official implementation for FlashKAT.
Hybrid ML + physics model of the Earth's atmosphere
PARADIS, a lightweight and flexible weather forecast model that tries to Keep It Simple.
This repository contains the SwinV2_Weather model, developed for the "Analyzing and Exploring Training Recipes for Large-Scale Transformer-Based Weather Prediction" paper. The repo includes trainin…
(ICLR 2022 Spotlight) Official PyTorch implementation of "How Do Vision Transformers Work?"
PDFM Embeddings: location-based vectors for geo-spatial analysis.
Lexcube: 3D Data Cube Visualization in Jupyter Notebooks
A collection of tricks and tools to speed up transformer models
A package to train machine learning models on geospatial data, mainly for weather and climate. Used to run ArchesWeather and ArchesWeatherGen
Implementation of the Aurora model for Earth system forecasting
Ai2 Climate Emulator: fast machine learning models for weather and climate prediction
[CVPR 2024] Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning
PDE-VAE: Variational Autoencoder for Extracting Interpretable Physical Parameters from Spatiotemporal Systems using Unsupervised Learning
River Water Segmentation in Close-Range Remote Sensing Imagery
A benchmark for the next generation of data-driven global weather models.
Implementation of the Prithvi WxC Foundation Model and Downstream Tasks