kerner-lab ORG

Code repository for EarthShift: Benchmarking the robustness of geospatial foundation models (GFMs) to realistic distribution shifts in Earth Observation

A benchmarking testbed that measures how well AI models trained on satellite imagery hold up when deployed under new conditions like different regions, sensors, seasons, or resolutions they never saw during training.

A codebase for pretraining multi-billion-scale sparse GPTs.

Mars-Bench is a standardized benchmark for evaluating vision models on Martian surface and orbital imagery, covering 20 datasets across classification, segmentation, and detection tasks.

No description.

Code repository for paper "Pretrain Where? Investigating How Pretraining Data Diversity Impacts Geospatial Foundation Model Performance". Code, data and model weights will be made public soon.

No description.

An All-MLP Sequence Modeling Architecture That Excels at Copying (Casual Relational Networks) Resources. https://arxiv.org/abs/2406.16168

Mars Orbital Model (MOMO)

ConeQuest: Open source data for Cone Segmentation on Mars

A one stop library for all major bias amplification metrics

No description.

No description.

A Semi-supervised Learning approach for Image segmentation of Crop field boundaries

No description.

No description.

Kerner Lab Website

CLI for fiboa (validation, inspection, schema and file creation, etc.)

Domain-agnostic Outlier Ranking Algorithms (DORA) - SMD cross-divisional use case demonstration of AI/ML