diff --git a/README.md b/README.md index 957e25672f8f6387684905bf498349c27fae2643..fa16a2c111a8dd95ae550ad20bcc784d43ac006d 100644 --- a/README.md +++ b/README.md @@ -55,6 +55,7 @@ Optional dependencies: interferometry responses) - [mpi4py](https://mpi4py.scipy.org) (for MPI-parallel execution) - [matplotlib](https://matplotlib.org/) (for field plotting) +- [pypocketfft](https://gitlab.mpcdf.mpg.de/mtr/pypocketfft) (for faster FFTs) ### Sources @@ -89,6 +90,14 @@ MPI support is added via: sudo apt-get install python3-mpi4py +Pypocketfft is added via: + pip3 install --user git+https://gitlab.mpcdf.mpg.de/mtr/pypocketfft + +If this library is present, NIFTy will detect it automatically and prefer +it over SciPy's FFT. The underlying code is actually the same, but +pypocketfft is compiled with optimizations for the host CPU and can provide +significantly faster transforms. + ### Running the tests To run the tests, additional packages are required: diff --git a/docs/source/installation.rst b/docs/source/installation.rst index 1836ffb0829ee67eeb98a61b0c2e51092b98ccec..0313bd7a669bbcbc8bdb073593de5c2a59894633 100644 --- a/docs/source/installation.rst +++ b/docs/source/installation.rst @@ -26,6 +26,14 @@ MPI support is added via:: sudo apt-get install python3-mpi4py +Pypocketfft is added via: + pip3 install --user git+https://gitlab.mpcdf.mpg.de/mtr/pypocketfft + +If this library is present, NIFTy will detect it automatically and prefer +it over SciPy's FFT. The underlying code is actually the same, but +pypocketfft is compiled with optimizations for the host CPU and can provide +significantly faster transforms. + NIFTy documentation is provided by Sphinx. To build the documentation:: sudo apt-get install python3-sphinx-rtd-theme dvipng