an ID associated with each each data point, i.e., each line in the data table.
a description of the experiment run
ranking information to be used in future auto generation of scoring functions
enabled/disabled state
The implementation
MUST support CSV import/export.
MUST be robust when loading data files with missing meta data fields
SHOULD support new custom fields without breaking old data
One solution for CSV support could be to include all meta data fields as separate columns in the CSV. This will force the meta data format to be only one level deep
Examples of meta data can be:
The implementation
One solution for CSV support could be to include all meta data fields as separate columns in the CSV. This will force the meta data format to be only one level deep