Installation

We recommend installing osl-dynamics using Miniforge.

Conda / Mamba Installation

Miniforge (conda/mamba) can be installed with:

curl -LO "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-$(uname)-$(uname -m).sh"
bash Miniforge3-$(uname)-$(uname -m).sh
rm Miniforge3-$(uname)-$(uname -m).sh

Linux Instructions

The following lines can be used to download a conda environment file and install osl-dynamics with its dependencies.

curl -LO https://raw.githubusercontent.com/OHBA-analysis/osl-dynamics/refs/heads/main/envs/osld-tf.yml
mamba env create -f osld-tf.yml
rm osld-tf.yml

If you have a GPU, then use the osld-tf-cuda.yml environment file:

curl -LO https://raw.githubusercontent.com/OHBA-analysis/osl-dynamics/refs/heads/main/envs/osld-tf-cuda.yml
mamba env create -f osld-tf-cuda.yml
rm osld-tf-cuda.yml

Mac Instructions

If you have an M-series (M1, M2, M3) chip, the following lines can be used to download a conda environment file and install osl-dynamics with its dependencies.

curl -LO https://raw.githubusercontent.com/OHBA-analysis/osl-dynamics/refs/heads/main/envs/osld-tf.yml
mamba env create -f osld-tf.yml
rm osld-tf.yml

If you have an Intel chip, then use the osld-tf-macos.yml environment file:

curl -LO https://raw.githubusercontent.com/OHBA-analysis/osl-dynamics/refs/heads/main/envs/osld-tf-macos.yml
mamba env create -f osld-tf-macos.yml
rm osld-tf-macos.yml

Windows Instructions

If you are using a Windows computer, we recommend first installing linux (Ubuntu) as a Windows Subsystem (WSL) by following the instructions here. Then follow the Conda/Mamba and Linux instructions above in the Ubuntu terminal.

Oxford-Specific Computers (hbaws, BMRC)

See the instructions on the GitHub README.

Test your GPUs are working

You can use the following to check if TensorFlow is using any GPUs you have available:

conda activate osld
python -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))"

This should print a list of the GPUs you have available (or an empty list [] if there’s none available).

Install without TensorFlow

If you want to install osl-dynamics without TensorFlow, then use the osld.yml environment file:

curl -LO https://raw.githubusercontent.com/OHBA-analysis/osl-dynamics/refs/heads/main/envs/osld.yml
mamba env create -f osld.yml
rm osld.yml

This environment file will only install the osl-dynamics package. You will have to install the dependencies (numpy, scipy, etc) yourself.

Install the latest development code (optional)

You should only need to do this if you need a feature or fix that has not been released on pip yet.

Once you have created the osld conda environment (see instructions above) you can install the latest development version on the GitHub repository with:

conda activate osld
pip install git+https://github.com/OHBA-analysis/osl-dynamics.git

Install the source code (optional)

Once you have created the osld conda environment (see instructions above) you can install a local copy of the source code (GitHub repository) into it.

git clone https://github.com/OHBA-analysis/osl-dynamics.git
conda activate osld
cd osl-dynamics
pip install -e .

Now you can run osl-dynamics with your own local changes to the code.

Removing osl-dynamics

To remove osl-dynamics simply delete the conda environment:

conda env remove -n osld
conda clean --all