either specify each filename, separated with commas, as in --data_filepath=datasets/batch/BUZZI/1/screening.csv,datasets/batch/BUZZI/2/screening.csv
or provide a regular expression to match folders and/or files, as in --data_filepath=datasets/batch/BUZZI/*/screening.csv
The first option seems more controlled as the exact list of files is provided. The second option is shorter to write but the regex can expand to uncontrolled files.
For now, option 1 is more reasonnable (and also considering that this script will be deployed in a real environment).
For proiding several inputs, we have two ways:
--data_filepath=datasets/batch/BUZZI/1/screening.csv,datasets/batch/BUZZI/2/screening.csv
--data_filepath=datasets/batch/BUZZI/*/screening.csv
The first option seems more controlled as the exact list of files is provided. The second option is shorter to write but the regex can expand to uncontrolled files.
For now, option 1 is more reasonnable (and also considering that this script will be deployed in a real environment).