Skip to content

Document Windows support; fix Conda instructions #558

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mihaic merged 1 commit into from
Dec 9, 2024
Merged

Document Windows support; fix Conda instructions #558

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 5 additions & 13 deletions README.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,23 +27,15 @@ package name), we use ``brainiak``.
Quickstart
==========

If you have `Conda <conda.io>`_::

conda install -c brainiak -c defaults -c conda-forge brainiak

Otherwise, or if you want to compile from source, install the requirements (see
`docs/installation`) and then install from PyPI::
You can install BrainIAK wheels from PyPI on Linux, macOS, and Windows, e.g. using pip::

python3 -m pip install brainiak

Note that to use the ``brainiak.matnormal`` package, you need to install
additional dependencies. As of October 2020, the required versions are not
available as Conda packages, so you should install from PyPI, even when using
Conda::
If you need MPI, see the installation requirements in `docs/installation`.

python3 -m pip install -U tensorflow tensorflow-probability
If you have `Conda <conda.io>`_, you can also use our Conda packages (not available for Windows). Installing BrainIAK will also install MPI::

Note that we do not support Windows.
conda install -c brainiak -c conda-forge brainiak


Docker
Expand All @@ -67,7 +59,7 @@ Note that we do not support MPI execution using Docker containers and that perfo
Support
=======

If you have a question or feedback, chat with us on `Gitter
If you have a question or feedback, chat with us on `Matrix via Gitter
<https://gitter.im/brainiak/brainiak>`_ or email our list at
brainiak@googlegroups.com. If you find a problem with BrainIAK, you can also
`open an issue on GitHub <https://github.com/brainiak/brainiak/issues>`_.
Expand Down
32 changes: 26 additions & 6 deletions docs/installation.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,27 +2,47 @@
Installation
============

Binaries
++++++++


PyPI
====

You can install BrainIAK wheels from PyPI on Linux, macOS, and Windows, e.g. using pip::

python3 -m pip install brainiak


MPI
---

On Linux and macOS, there are no MPI4PY binaries in PyPI, so you need to install MPI separately, e.g., following the BrainIAK `source installation instructions <#requirements>`_.
Copy link
Contributor

@davidt0x davidt0x Dec 9, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You can get wheels for mpi4py on linux and macos (see here). You need to add the extra index argument though. Some on linux, a one-line install looks like this:

pip install brainiak[matnormal]==0.12.a5 impi-rt --extra-index-url https://pypi.anaconda.org/mpi4py/simple

I just tested this in WSL and it worked for me, all tests passed. For windows, its even simpler since there are mpi4py wheels already on PyPI. Just this works:

pip install brainiak[matnormal]==0.12.a5 impi-rt

For MacOS, you can get the wheels similar to linux but I believe you still need to brew install or macports install opemmpi or mpich.


On Windows, MPI4PY binaries will be installed as BrainIAK dependencies. You can also install an MPI implementation available on PyPI such as Intel MPI::

python3 -m pip install impi-rt


Conda
+++++
=====

Make sure you have the ``conda-forge`` channel active, because some of our
dependencies are not available in the default channels.

Use our ``brainiak`` channel to install::

conda install -c brainiak -c defaults -c conda-forge brainiak
conda install -c brainiak -c conda-forge brainiak

Note that `~brainiak.funcalign.sssrm.SSSRM` currently uses Theano, which
requires the Xcode Command Line Tools on `MacOS`_.

Source
++++++

Requirements
============

We support Linux and MacOS with Python version 3.5 or higher. Most of the
dependencies will be installed automatically. However, a few need to be
We support Linux, MacOS, and Windows with Python version 3.9 or higher. Most of
Copy link
Contributor

@davidt0x davidt0x Dec 9, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We don't support 3.13 at the moment, tensorflow is the problem I think.

the dependencies will be installed automatically. However, a few need to be
installed manually.


Expand Down
Loading