v1.0.3

Packaging

• split [torch] optional dependencies into [torch-cpu] (includes finufft) and [torch-cuda] (includes cufinufft) in order to avoid pip install error on windows systems

Documentation

• updated installation instructions

v1.0.2

Code

• minor fix for processing.eprfbp2d and processing.eprfbp3d functions (displayer=None was not working as expected)

• changed normalization in processing.eprfbp2d and processing.eprfbp3d to get a consistent quadrature scheme

Documentation

• added the mathematical description of the filtered backrojection scheme implemented in the package

• fixed Gaussian derivative normalization factor for demo with simulated reference spectra

v1.0.1

Code

• temporary fix related to a multithreading issue with FINUFFT (see FINUFFT issue #596): introduced a decorator in backends.py to change the default value of the nthreads keyword argument of the finufft functions according to the number of physical cores (or the OMP_NUM_THREADS environment variable if set)

• increased to 1E6 the default maximal number of iterations (parameter nitermax) for optimization schemes and related functions

• fixed type inference for backend.from_numpy() (torch backend)

• fixed typo in function name (read_bruker_best3_dataset instead of read_bruker_bes3t_dataset), old name was kept available for backward compatibility

• fixed sphinx rendering issues in various function headers in multisrc.py

Documentation

• updated installation instructions to allow cupy installation using pip

• fixed bibtex reference [Bar21]

• fixed minor issues in demonstration examples

• fixed pip installation instructions in README.md

v1.0.0

First public release