brainiak / brainiak.github.io

https://brainiak.org

Brain Imaging Analysis Kit

The Brain Imaging Analysis Kit is a package of Python modules useful for neuroscience, primarily focused on functional Magnetic Resonance Imaging (fMRI) analysis.

The package was originally created by a collaboration between Intel and the Princeton Neuroscience Institute (PNI).

To reduce verbosity, we may refer to the Brain Imaging Analysis Kit using the BrainIAK abbreviation. Whenever lowercase spelling is used (e.g., Python package name), we use brainiak.

Quickstart

If you have Conda:

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:

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:

python3 -m pip install -U tensorflow tensorflow-probability

Note that we do not support Windows.

Docker

You can also test BrainIAK without installing it using Docker:

docker pull brainiak/brainiak
docker run -it -p 8888:8888 -v brainiak:/mnt --name demo brainiak/brainiak

To run Jupyter notebooks in the running container, try:

python3 -m notebook --allow-root --no-browser --ip=0.0.0.0

Then visit http://localhost:8888 in your browser and enter the token. Protip: run screen before running the notebook command.

Note that we do not support MPI execution using Docker containers and that performance will not be optimal.

Support

If you have a question or feedback, chat with us on Gitter or email our list at [email protected]. If you find a problem with BrainIAK, you can also open an issue on GitHub.

Examples

We include BrainIAK usage examples in the examples directory of the code repository, e.g., funcalign/srm_image_prediction_example.ipynb.

To run the examples, download an archive of the latest BrainIAK release from GitHub. Note that we only support the latest release at this moment, so make sure to upgrade your BrainIAK installation.

Documentation

The documentation is available at http://brainiak.org/docs.

Contributing

We welcome contributions. Have a look at the issues labeled "easy" for starting contribution ideas. Please read the guide in CONTRIBUTING.rst first.

Citing

Please cite BrainIAK in your publications as: "Brain Imaging Analysis Kit, http://brainiak.org." Additionally, if you use RRIDs to identify resources, please mention BrainIAK as "Brain Imaging Analysis Kit, RRID:SCR_014824". Finally, please cite the publications referenced in the documentation of the BrainIAK modules you use, e.g., SRM.

brainiak.github.io's Issues

Add hackathons to events

Explain analyses more on the front page

Tutorial Repo update

@danielsuo We need to update all instances of the tutorial repo on the install page:
git clone https://github.com/brainiak/brainiak-tutorials.git

I made a mistake in the current URL.

Create default template

Add docs

Add favicon

Wrong plot title in documentation!

In this https://brainiak.org/events/ohbm2018/brainiak_sample_tutorials/06-rsa.html of your official webpage you converted the RSM(Representational Similarly Matrix) sample to RDM(Representational Dissimilarity Matrix) and plotted it while wrongly you wrote RSM as the title of the plot. Just see the cell number 12! (The title should be "RDM, sorted, rFFA" while it is "RSM, sorted, rFFA" as the previous cell!).

Tiny mistake but confused me a lot!

Thanks for your consideration in advance!

Add a way to keep track of citing papers automatically

This will not be 100% accurate, so we could add a disclaimer and present it after the curated list of papers we present.

./bin/install check if brew / bundle installed

Conda install instructions

On the brainiak homepage http://brainiak.org, instructions for conda install have not been added under "Getting Started".

Nice page scrolling navigation bar

Example here)

Add "Need Help" button at the top of the brainiak homepage

Hi,

Currently the "Need Help" button is embedded towards the bottom of the brainiak Home page. It would make it easier for users to know where to log issues etc. to see the link on top.

Thanks,
Manoj

Add author tag in publications

Update footer links

Add BrainIAK change log

Add events

• neuroscience 2017
• rtfun2017

fix weird scrolling issue at the bottom in mobile

It seems like the scrolling gets “stuck”

Fix hamburger menu icon

Add publications

Add course / learning materials

Change hero image in redesign2

Should we add the logos of participating institutions?

Right now, Princeton, Yale, Northeastern?

At least a few others in the orbit, but maybe not directly involved (e.g., VT, UTA, Chicago, UPenn Hospital)

Add publications to publications page

Automate docs uploading

Currently we are making regular pull requests, which are difficult to review.

Add HTTPS

HTTPS is now supported by GitHub:
https://blog.github.com/2018-05-01-github-pages-custom-domains-https/

Current RSA tutorial does not reflect BRSA

I think this should be at least declared in the tutorial, in case people who are looking for BRSA example get confused.

Add link to GitHub repository

Add team page

Cluster environment description update

A naive student tried to get started on the cluster environment and had trouble so here are some recommendation of how to change the get started text.

The tabs for 'Cluster' and 'Cluster (admin)' should be renamed to something like 'Cluster (personal only)' and 'Cluster (shared)' and then each has a description at the top like:

"

Cluster (personal only)

Use these instructions if you wish to create a personal environment on a cluster. You will set a conda environment that you have access to and can use for running these tutorials. You are expected to have knowledge of SLURM parameters, your cluster, to launch the Jupyter notebooks and run batch jobs. Your cluster administrator can help you with setting the SLURM parameters. If other people have already set up these tutorials or intend to, it might be better to use the instructions for 'Cluster (shared)'.
"

"

Cluster (shared)

Use these instructions to set up an environment that can be used multiple users on a cluster. Included are instructions for an admin to create a shared environment and the users need to change some of the tutorial scripts to accommodate This will likely involve creating modules that can be loaded by other users and downloading data to a shared location. You are expected to have knowledge of SLURM parameters, your cluster, to launch the Jupyter notebooks and run batch jobs. Your cluster administrator can help you with setting the SLURM parameters. If you only wish to set up the tutorials for yourself, it might be better to use the instructions for 'Cluster (personal use)'
"

In the current 'Cluster' tab it lists the same instructions twice:

"Check your environment
Check file setup_environment.sh for CONDA_ENV=mybrainiak
Check configuration='cluster'"

It would be good to have a line added to the current 'Cluster' and 'Cluster (admin)' tabs say something like:

"When running the 'run_jupyter_remote_server.sh' script, two problems often arise.

1. if you submit this as a job you may get an issue where the script cannot be run interactively. If so, change the first line of this file to read: #!/bin/bash
2. depending on your cluster configuration, your server name may be inappropriately referenced. This will mean that the suggested host name this script outputs will be wrong (typically because it just finds the name of the node you are on). In such cases, you can manually edit the server name in the setup_environment.sh script."

Why do we tell them to use run_jupyter_remote_server when launching on a cluster? Why not use the run_jupyter_remote_cluster.sh for this use? The only difference is using MPI