search

incf-nidash / nidmresults-fsl

A python library to export FSL's feat results to NIDM-Results
http://nidm.nidash.org/specs/nidm-results.html
MIT License
3 stars 10 forks source link

Add possibility to read user-specified info from a config file #144

Open cmaumet opened 5 years ago

cmaumet commented 5 years ago

Discussed with @mih today.

If the user-specific metadata (groups, space) is not provided in the command-line or as argument, look for a file in a specific location, e.g. nidm_metadata.json, that will contain that info.

mih commented 5 years ago

It is actually not sufficient to have a config file in order to enable automated processing. If I do not specify --groups for a higher-level analysis the command fails. If I do specific --groups for a 1st-level analysis the command also fails. That means I need knowledge about an analysis, before I can have a look at its metadata -- a bit ironic ;-)

satra commented 5 years ago

@mih and @cmaumet - shouldn't nidmresults be able to figure out what kind of a feat analysis was run from an introspection of the directories (including potential repeated analysis) ?

cmaumet commented 5 years ago

@mih, @satra: those metadata cannot be automatically extracted from the software outputs that's why we had to make them mandatory for the user to provide. I understand this is a problem in the context of using the exporter on already computed analyses for which the person doing the conversion does not have the knowledge about the analysis... Maybe something we could relax in NIDM-Results, but these were info that we strongly felt had to be included for the packs to be re-usable...