neuropsych_mtl

Accompanying package: miniMTL

Directory Structure

• bash_scripts
  • SLURM submission scripts for running parts of this code
    • Mostly for narval (compute canada)
• conf_balancing
  • Scripts related to balancing datasets
  • general_class_balancer.py
    • General Class Balancer (GCB) implementation from mleming
  • mleming_cv_splits.py
    • Using GCB to make balanced test sets
  • mleming_datasets.py
    • Using GCB to make datasets
  • pymatch_cv_splits.py
    • Using propensity score matching with pymatch to make balanced test sets
• datasets
  • .txt files for each condition with ids for case/control to include in dataset
  • cv_folds
    • .csv files for dataset & ids in each fold for balanced test set cross validation
    • gen_cv_set_figures.py
      • Plots for the balanced test sets
  • gen_dataset_figures.py - Plots for the datasets
  • figures
• ML
  • benchmark_1
    • results
    • benchmark_1.py
    • benchmark_1_plots.ipynb
  • benchmark_2
    • results
    • benchmark_2.py
    • benchmark_2_plots.ipynb
  • benchmark_2
    • TODO
• MTL
  • TODO
• stats
  • results
  • dataset_exploration.ipynb
    • Plots & tables
  • generate_betamaps.py
    • script to generate output of Connectome Wide Association Studies (CWAS)
  • mtd_cv.py
    • Script for cross-validation of effect size (MTD)
  • mtd_stats.py
    • Script to generate effect sizes, p-values, confidence intervals (CIs), and count what survived FDR
  • test_set_chance_level.ipynb
    • Notebook to check the chance level in each dataset.
  • util.py