Skip to content

lebel.md

Latest commit

 

History

History
29 lines (24 loc) · 948 Bytes

lebel.md

File metadata and controls

29 lines (24 loc) · 948 Bytes

LeBel Setup

This guide describes how to prepare LeBel data for LITcoder experiments.

Requirements

  • fMRIPrep preprocessed BOLD in MNI152NLin2009cAsym at 2 mm (res-2)
  • Post-processing script: fmri_processing/to_mni_lebel.py
  • Dataset-level files placed in data/lebel/neural_data/:
    • grids_huge.jbl
    • trfiles_huge.jbl
  • Per-subject pickles placed directly in data/lebel/neural_data/:
    • noslice_sub-<SUBJECT_ID>_story_data.pkl
    • noslice_sub-<SUBJECT_ID>_story_data_surface.pkl

Steps

  1. Run fMRIPrep as usual.
  2. Run the post-processing script:
python fmri_processing/to_mni_lebel.py
  1. Move the produced per-subject pickle files into data/lebel/neural_data/.

Expected Locations

<repo_root>/data/lebel/neural_data/lebel_data.pkl
<repo_root>/data/lebel/neural_data/noslice_sub-<SUBJECT_ID>_story_data.pkl
<repo_root>/data/lebel/neural_data/noslice_sub-<SUBJECT_ID>_story_data_surface.pkl