HTTomo is a Python framework for high-performance tomographic data processing mainly targeting GPU-compute. It orchestrates distributed I/O and CPU/GPU workflows using MPI, while providing YAML-based access to processing methods from libraries such as TomoPy and HTTomolibgpu.
The HTTomo documentation contains installation instructions, a quickstart, ready-to-use pipelines, reference material and developer guidance.
After installation, use python -m httomo --help to inspect the command
line interface. A typical workflow is to validate a pipeline and then run it:
python -m httomo check pipeline.yaml input.nxs
python -m httomo run input.nxs pipeline.yaml output_directoryWe use the setuptools-git-versioning
package for automatically determining the version from the latest git tag.
For this to work, release tags should start with a v followed by the actual version,
e.g. v1.1.0a.