ML competitions

Winning solutions in Geospatial and Earth observation

Each competition keeps placement-level public sources and a source image linked back to the competition provider.

View all matching competition records

Named public records

Inspect the evidence itself

These records are curated examples from the detailed snapshot. Their count can be lower than the broader source count above because the sources are collected and refreshed independently.

Exact public-code imports

Each named record identifies the context of an exact import. An import can appear in source, an example, a test, documentation, or a notebook.

19

View 19 named public recordsExpand

Import in source

raster-vision

Open-source framework for deep learning on satellite and aerial imagery.

Import in source

solaris

Geospatial machine-learning processing framework built for SpaceNet challenges.

Import in source

DLR-MF-DAS/SmallMinesDS

Benchmarking foundation models for semantic segmentation of small scale mining activities

Import in source

IBM/peft-geofm

Parameter-Efficient Fine-Tuning for Geospatial Foundation Models

Import in source

IBM/TerraMesh-Masks

TerraMesh-Masks provides training and evaluation dataset for open-vocabulary segmentation of satellite imagery.

Import in source

terratorch

PyTorch toolkit for fine-tuning Earth-observation foundation models.

Import in example

IBM/vllm

vLLM with support for span semantics

Import in source

prithvi-pytorch

NASA-IBM Earth-observation foundation models.

Import in example

PaddlePaddle/PaddleScience

PaddleScience is SDK and library for developing AI-driven scientific computing applications based on PaddlePaddle.

Import in source

GeoSeg

Toolbox for semantic segmentation of remote-sensing imagery.

Import in source

zhu-xlab/GCP

Official implementation of "Global Collinearity-aware Polygonizer for Polygonal Building Mapping in Remote Sensing"

Import in source

zhu-xlab/LandSegmenter

This repository contains the official implementation of the paper "LandSegmenter: Towards a Flexible Foundation Model for Land Use and Land Cover Mapping".

Research citations

A citation establishes research use or discovery. It does not by itself prove a dependency, deployment, endorsement, or commercial license.

241

View 241 named public recordsExpand

Verified Hugging Face artifacts

Each record has public code-use or runtime-dependency evidence. Reference-only cards are not shown here.

27

View 27 named public recordsExpand

Verified runtime dependency · Hugging Face space

antan123/SolarIQ

0 public downloads · python ast import in app.py

Verified runtime dependency · Hugging Face space

caixiaoshun/cloudseg

0 public downloads · python ast import in app.py

Verified public-code use · Hugging Face space

HebaAllah/depi_final

0 public downloads · python ast import in app.py

Verified runtime dependency · Hugging Face space

msradam/riprap-vllm

0 public downloads · dependency manifest in Dockerfile

Verified runtime dependency · Hugging Face space

shuklaabhinavv/roadx

0 public downloads · dependency manifest in Dockerfile

Inspect all public GitHub evidence