NRCan/geo-deep-learning

Deep learning applied to georeferenced datasets

59
/ 100
Established

This project helps geoscientists and remote sensing specialists train deep learning models using various Earth observation data sources like satellite imagery. It takes in multi-sensor geospatial datasets, often in WebDataset or CSV format, and outputs trained models for tasks such as identifying land cover, detecting objects, or performing regressions on geographic features. This tool is for researchers and practitioners who need to analyze complex spatial data.

193 stars.

Use this if you are working with Earth observation data from multiple sensors and need a flexible framework to train deep learning models for tasks like semantic segmentation or object detection.

Not ideal if you need a simple, out-of-the-box solution for common image processing tasks that don't involve complex geospatial or multi-sensor data.

remote-sensing geospatial-analysis earth-observation satellite-imagery environmental-monitoring
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 23 / 25

How are scores calculated?

Stars

193

Forks

66

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 09, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NRCan/geo-deep-learning"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.