Home

Welcome to Thunderseg !

Thunderseg is an open-source package built on top of pytorch and pytorch-lightning. This package focus on deep learning image segmentation tasks for remote sensing applications, including vegetation, cropland, artificial structures, etc.

The goal of this package is to collect the state-of-art image segmentation methods and provide painless image segmentation workflow to the users.

This document will assume you have fundimental understanding of:

• Python
• Linux CLI
• Conda
• Docker (Optional)
• Apptainer (Optional)

You may click these hyperlinks above to access corresponding tutorial.

System Requirement

Thunderseg should be able to run on any Linux-based system. The package has been tested in Ubuntu 22.04.4 LTS running under WSL2.

Test environment:

Category	Model
CPU	AMD Ryzen 7 7840HS
GPU	Nvidia 4070 Super
Memory	64GB DDR5
Operate System	Windows 10 Pro N 64-bit WSL2

Need Help?

Meanwhile, if you have any question relate to the package or the documentation, please feel free to submit , join , or send me