README.md

ANTsTorch documentation

ANTsTorch provides PyTorch implementations of ANTsX deep-learning architectures and applications, together with differentiable image-processing and registration tools.

Getting started

• Installation and project overview
• Available applications
• Self-contained ANTsX examples

Registration and image processing

• SyN registration tutorial
• N4 and diffeomorphic B-spline registration tutorial

Normalizing flows

• Hybrid LAMNR flow trainer

Validation and development

• Model validation and ANTsPyNet parity
• Development tools
• Application verification scripts

The public Python API is exposed through the antstorch package. Application functions are available from the top-level namespace, while registration and benchmark functionality is organized under antstorch.syn, antstorch.bspline_flows, and antstorch.benchmark.

Documentation website

• Site contents
• Runnable examples
• Build locally and publish on Read the Docs