Details
-
Improvement
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
1.1.0
-
None
Description
I sat down with huikyole and we agreed on the following improvements to DatasetLoader:
1) Stop distinguishing between reference and target datasets. This was only necessary for certain metrics, but in our mission to make the process of dataset loading more flexible we thought it would be better to keep things as general as possible at this step. Furthermore, we would also like to make it possible to process multiple observation datasets at once.
2) Set default loader to local. By default, if a data_source isn't specified, a KeyError is thrown. This will be changed to using the load_multiple_files loader from local. Furthermore, we do not need to specify 'local_multiple' and 'local' separately since the latter loader can easily handle the single file case, so this will be the new default loader rather than local.load_file.
3) Change 'data_source' keyword to 'loader_name'. This needs to be done at the config file level too.
Attachments
Issue Links
- links to
