Python code for fast MBIR (Model Based Iterative Reconstruction)
This is a python wrapper for High Performance Imaging's supervoxel C code, HPImaging/sv-mbirct.
Full documentation is available at: https://svmbir.readthedocs.io
Currently supporting Python 3.7-3.10, on MacOS and Linux (Windows possible but not actively maintained).
svmbir packages are available from conda-forge and PyPI, or can be built and installed from source.
- (recommended) Create a clean virtural environment, such as
conda create -n svmbir python=3.8
conda activate svmbir
- To install from conda-forge,
conda install -c conda-forge svmbir
- To install from PyPI,
pip install svmbir
- Installing from source (requires GNU/gcc compiler, OMP libraries),
# In top repository folder,
CC=gcc pip install . # also supports Intel "icc"
See here for more details.
- Download demo.zip at https://github.com/cabouman/svmbir/blob/master/demo.zip.
- Uncompress the zip file and change into demo folder.
- In your terminal window, install required dependencies of demo.
pip install -r requirements_demo.txt
- In your terminal window, use python to run each demo.