Skip to content

Commit

Permalink
Doc: convert files to markdown
Browse files Browse the repository at this point in the history
  • Loading branch information
@paugier
paugier committed Jan 24, 2024
1 parent 9230ab5 commit a9c7d29
Show file tree
Hide file tree
Showing 37 changed files with 1,396 additions and 1,783 deletions.
68 changes: 25 additions & 43 deletions CHANGES.rst → CHANGES.md
View file
Original file line number Diff line number Diff line change
@@ -1,109 +1,91 @@
0.3.5 (2023-09-20)
------------------
# Release notes

## 0.3.5 (2023-09-20)

- Support for Numpy 1.26

0.3.4 (2023-01-05)
------------------
## 0.3.4 (2023-01-05)

- Support for Numpy 1.24 and Python 3.11
- Internal: switch to `src layout
<https://packaging.python.org/en/latest/discussions/src-layout-vs-flat-layout/>`__
and to `Nox <https://nox.thea.codes>`__

0.3.3 (2022-09-13)
------------------
- Internal: switch to [src layout](https://packaging.python.org/en/latest/discussions/src-layout-vs-flat-layout/)
and to [Nox](https://nox.thea.codes)

- Support for `editable_wheel (PEP 660) <https://peps.python.org/pep-0660/>`__ and setuptools>=65.3
## 0.3.3 (2022-09-13)

- Support for [editable_wheel (PEP 660)](https://peps.python.org/pep-0660/) and setuptools>=65.3
- CI on Github Actions for Linux, Windows and macOS

- Support for Python 3.8, 3.9 and 3.10

0.3.2 (2020-10-15)
------------------
## 0.3.2 (2020-10-15)

- Few optimizations in 3D operators
- pyproject.toml and isolated build
- Various bugfixes (in particular, problems with `-Wl,--as-needed`)

0.3.1 (2020-05-09)
------------------
## 0.3.1 (2020-05-09)

- Various bugfixes and improvements + compatibility Python 3.8

0.3.0 (2019-09-21)
------------------
## 0.3.0 (2019-09-21)

- classes using mpi4py-fft
- compatibility transonic 0.4.0

0.2.9 (2019-03-07)
------------------
## 0.2.9 (2019-03-07)

- Minimal support for Windows
- Slightly better logging of bad parameters
- Optimization with dealiasing coeff for fft3dmpi_with_fftw1d

0.2.8 (2019-01-27)
------------------
## 0.2.8 (2019-01-27)

- ``pip install`` from an empty environment should works
- `pip install` from an empty environment should works
- Use transonic
- ``compute_spectrum_kykx`` (in 2d operators)
- `compute_spectrum_kykx` (in 2d operators)
- Supports Python >= 3.6

0.2.7 (2018-10-22)
------------------
## 0.2.7 (2018-10-22)

- Bugfix setup without mpicxx OpenMPI

0.2.5 (2018-07-20)
------------------
## 0.2.5 (2018-07-20)

- Bugfix classes using fftw_mpi (see `issue #14
<https://foss.heptapod.net/fluiddyn/fluidfft/issues/14>`_)
- Bugfix classes using fftw_mpi (see [issue #14](https://foss.heptapod.net/fluiddyn/fluidfft/issues/14))
- New function mean_global

0.2.4 (2018-07-01)
------------------
## 0.2.4 (2018-07-01)

- bugfixes and code improvements
- support PyPy3 and macOS
- More robust to lack of pyfftw and to mkl
- Support fft="default" in operator classes

0.2.3 (2018-06-04)
------------------
## 0.2.3 (2018-06-04)

- Less bugs
- Pypy compatibility
- Support jinja2 as fallback for mako
- rotfft_from_vecfft_outin, 3d scatter/gather and compute spectra routines

0.2.2
-----
## 0.2.2

- Less bugs
- Install on clusters
- Better checks for bad input parameters
- Operator div_vv_fft_from_v
- FLUIDDYN_NUM_PROCS_BUILD to install on small computers

0.2.1
-----
## 0.2.1

- install with pip (+ configure install setup with ``~/.fluidfft-site.cfg``)
- install with pip (+ configure install setup with `~/.fluidfft-site.cfg`)

0.2.0
-----
## 0.2.0

- Much cleaner, much less bugs, better unittests
- Much faster building
- Much better operators 2d and 3d
- Tested on different clusters

0.1.0
-----
## 0.1.0

- fluidfft can be used with mpi in fluidsim.
131 changes: 131 additions & 0 deletions README.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,131 @@
![FluidFFT](https://foss.heptapod.net/fluiddyn/fluidfft/-/blob/branch/default/doc/logo.svg)
===========================================================================================

*Efficient and easy Fast Fourier Transform for Python*

[![Latest version](https://img.shields.io/pypi/v/fluidfft.svg)](https://pypi.org/project/fluidfft/)
![Supported Python versions](https://img.shields.io/pypi/pyversions/fluidfft.svg)
[![Documentation status](https://readthedocs.org/projects/fluidfft/badge/?version=latest)](http://fluidfft.readthedocs.org)
[![Code coverage](https://codecov.io/gh/fluiddyn/fluidfft/branch/branch%2Fdefault/graph/badge.svg)](https://codecov.io/gh/fluiddyn/fluidfft)
[![Heptapod CI](https://foss.heptapod.net/fluiddyn/fluidfft/badges/branch/default/pipeline.svg)](https://foss.heptapod.net/fluiddyn/fluidfft/-/pipelines)
[![Github Actions Linux](https://github.com/fluiddyn/fluidfft/actions/workflows/ci-linux.yml/badge.svg?branch=branch/default)](https://github.com/fluiddyn/fluidfft/actions/workflows/ci-linux.yml)
[![Github Actions Windows](https://github.com/fluiddyn/fluidfft/actions/workflows/ci-windows.yml/badge.svg?branch=branch/default)](https://github.com/fluiddyn/fluidfft/actions/workflows/ci-windows.yml)
[![Github Actions macOS](https://github.com/fluiddyn/fluidfft/actions/workflows/ci-macos.yml/badge.svg?branch=branch/default)](https://github.com/fluiddyn/fluidfft/actions/workflows/ci-macos.yml)

Fluidfft provides C++ classes and their Python wrapper classes written
in Cython useful to perform Fast Fourier Transform (FFT) with different
libraries, in particular

- [fftw3](http://www.fftw.org/) and
[fftw3-mpi](http://www.fftw.org/fftw3_doc/Distributed_002dmemory-FFTW-with-MPI.html)
- [pfft](https://github.com/mpip/pfft)
- [p3dfft](https://github.com/sdsc/p3dfft)
- [mpi4py-fft](https://bitbucket.org/mpi4py/mpi4py-fft)
- [cufft](https://developer.nvidia.com/cufft) (fft library by CUDA
running on GPU)

[pfft](https://github.com/mpip/pfft),
[p3dfft](https://github.com/sdsc/p3dfft) and
[mpi4py-fft](https://bitbucket.org/mpi4py/mpi4py-fft) are specialized in
computing FFT efficiently on several cores of big clusters. The data can
be split in pencils and can be distributed on several processes.

**Documentation**: <https://fluidfft.readthedocs.io>

Getting started
---------------

To try fluidfft without installation:
[![Binder notebook](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/fluiddyn/fluidfft/branch%2Fdefault?urlpath=lab/tree/doc/ipynb)

For a **basic installation** which relies only on a `pyFFTW` interface;
or provided you have the optional FFT libaries, that you need, installed
and discoverable in your path (see environment variables `LIBRARY_PATH`,
`LD_LIBRARY_PATH`, `CPATH`) it should be sufficient to run:

pip install fluidfft [--user]

Add `--user` flag if you are installing without setting up a virtual
environment.

Installation
------------

To take full advantage of fluidfft, consider installing the following
(optional) dependencies and configurations before installing fluidfft.
Click on the links to know more:

1. OpenMPI or equivalent
2. FFT libraries such as MPI-enabled FFTW (for 2D and 3D solvers) and
P3DFFT, PFFT (for 3D solvers) either using a package manager or
[from
source](https://fluidfft.readthedocs.io/en/latest/install/fft_libs.html)
3. Python packages `fluiddyn mako cython pyfftw pythran mpi4py`
4. [A C++11 compiler and BLAS
libraries](https://github.com/serge-sans-paille/pythran#installation)
and
[configure](https://fluidfft.readthedocs.io/en/latest/install.html#dependencies)
`~/.pythranrc` to customize compilation of Pythran extensions
5. [Configure](https://fluidfft.readthedocs.io/en/latest/install.html#basic-installation-with-pip)
`~/.fluidfft-site.cfg` to detect the FFT libraries and install
`fluidfft`

**Note**: Detailed instructions to install the above dependencies using
Anaconda / Miniconda or in a specific operating system such as Ubuntu,
macOS etc. can be found
[here](https://fluiddyn.readthedocs.io/en/latest/get_good_Python_env.html).

### C++ API

See a [working minimal example with
Makefile](https://fluidfft.readthedocs.io/en/latest/examples/cpp.html)
which illustrates how to use the C++ API.

Tests
-----

From the root directory:

make tests
make tests_mpi

Or, from the root directory or any of the \"test\" directories:

pytest -s
mpirun -np 2 pytest -s

How does it work?
-----------------

Fluidfft provides classes to use in a transparent way all these
libraries with an unified API. These classes are not limited to just
performing Fourier transforms. They are also an elegant solution to
efficiently perform operations on data in real and spectral spaces
(gradient, divergence, rotational, sum over wavenumbers, computation of
spectra, etc.) and easily deal with the data distribution (gather the
data on one process, scatter the data to many processes) without having
to know the internal organization of every FFT library.

Fluidfft hides the internal complication of (distributed) FFT libraries
and allows the user to find (by benchmarking) and to choose the most
efficient solution for a particular case. Fluidfft is therefore a very
useful tool to write HPC applications using FFT, as for example
pseudo-spectral simulation codes. In particular, fluidfft is used in the
Computational Fluid Dynamics (CFD) framework
[fluidsim](http://fluidsim.readthedocs.org).

License
-------

Fluidfft is distributed under the
[CeCILL](http://www.cecill.info/index.en.html) License, a GPL compatible
french license.

Metapapers and citations
------------------------

If you use FluidFFT to produce scientific articles, please cite our
metapapers presenting the [FluidDyn
project](https://openresearchsoftware.metajnl.com/articles/10.5334/jors.237/)
and
[Fluidfft](https://openresearchsoftware.metajnl.com/articles/10.5334/jors.238/):
Loading

0 comments on commit a9c7d29

Please sign in to comment.