Source code for paper: Decoding multi-joint hand movements from brain signals by learning a synergy-based neural manifold

How to Run: Instructions for Users

1. Download the Dataset

Download the dataset and place it in the Dataset directory. The dataset is organized into the following three folders:

• NeuralData: Contains invasive brain signal recordings from each session.
• KinData: Contains joint angle trajectories for 11 hand movements.
• Info: Contains session metadata.

2. Modify Configuration Files

In the config directory, you need to update the following configuration files:

• Data-related Config:
  In the datasets section, update the root_dir and raw_root_dir to the paths where your dataset is stored.

• Training-related Config:
  In the train section, update the root_dir and raw_root_dir to match the location of your dataset. You can also modify other training parameters in this file.

• Bayesian Hyperparameter Optimization Config:
  In the ray section, update the following parameters:

  • default_config_dir: Specifies the path to the configuration file.
  • exp_config: The filename (.yaml) that specifies the default configuration for training and data processing.
  • tune_hp_json: The filename (.json) that defines the hyperparameter search space.

3. Run the Code

To execute the code, use the main.py script. Before running it, ensure you update the Ray configuration name based on the session you are working with.

For example:

main("SynergyNet_ray_SBP220627S1")

This runs the SynergyNet model with neural signals (SBP) collected on 2022-06-27 from Session 1. The search space is defined in the file located at config/ray/SynergyNet_ray_SBP220627S1.yaml