howto

The pipeline can be ran as follows:

Rscript --vanilla ~/mini_ml/config_ml.R --data=/location/of/data --number_repeats=5 --number_folds=10 --learning_method=glmnet --preprocessing=mad_prune_features --n_parallel_cores=20 --out=glmnet.yml --learning_type=regression
Rscript --max-ppsize=500000 --vanilla ~/mini_ml/run_ml.R glmnet.yml
Rscript --max-ppsize=500000 --vanilla ~/mini_ml/analyze_results.R

• The first of the commands above prepares a configuration file (in this case to run a glmnet regression)
• The second actually runs the ML method
• The third is used once all methods in a given folder are successfully completed