• Dawn Finzi 📧
• Eshed Margalit
• Kendrick Kay*
• Daniel L. K. Yamins*
• Kalanit Grill-Spector*

*co-senior authors

This repository has the following components:

• notebooks/: Jupyter notebooks (saved as Markdown files) that reproduce all figures and statistics in the paper
• scripts/: standalone scripts to be run from the command line. Includes example scripts to run the mapping algorithm (fitting_one_to_one_unit2voxel.py and fitting_one_to_one_voxel2voxel.py among others).
• spacestream/: installable Python package with functionality including model feature extraction, model-to-brain mapping methods, task transfer, spatial and functional metrics of correspondence to the brain, and more.

To install the repository, follow these steps:

1. Create a virtual environment for this project (for example): python3.7 -m venv spacestream_env
2. Activate the environment: source spacestream_env/bin/activate
3. Clone the repository: git clone https://github.com/dawnfinzi/spacestream.git
4. Change your working directory: cd spacestream
5. Install the required packages: pip install -e .