Skip to content

Comparing latent space representations using autoencoders and vision transformers using fMRI data.

Notifications You must be signed in to change notification settings

JeffTheNinja57/research_workshop

Repository files navigation

Research Workshop CSAI 2024 - Syntax Sentinels group

Overview

This repository contains the code for the fMRI data processing described in our research paper. The project focuses on preprocessing fMRI data to generate embeddings suitable for machine learning models, particularly autoencoders.

Prerequisites

Install the required packages using the provided env_rw24.yml file:

conda env create -f env_rw24.yml
conda activate research_workshop

Setup

Check for Updates

git fetch
git pull origin develop

Running fMRI Preprocessing

Run the scripts in the following order:

  1. file_modifications.py: Handles necessary file modifications.
    python file_modifications.py
  2. fmri_data_processing.py: Main script for processing fMRI data.
    python fmri_data_processing.py
    Run multiprocessor_image_processing.py if you use a Mac, for faster processing speeds The preprocessed fMRI images will look like this: fMRI_sub-CSI1_ses-1_run-1_timestep-1
  3. final_ae.py: train the model on the images
python final_ae.py

About

Comparing latent space representations using autoencoders and vision transformers using fMRI data.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages