You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the current implementation experiments are only defined by name and they have the possibility of one string for description. Things get very inconvenient though when many experiments are included, particularly in the main index screen.
I would therefore suggest that the model gets modified, in order to allow more fields for every experiment entry. In particular, one can add "source language", "target language", "dataset name". This can be specified in the medatada file of the imported folder, or also be modified after the importing.
Additionally, the possibility of free tagging based on an additional sql table should be possible.
This fields can be used in a future implementation for sorting/filtering the main screen
The text was updated successfully, but these errors were encountered:
In the current implementation experiments are only defined by name and they have the possibility of one string for description. Things get very inconvenient though when many experiments are included, particularly in the main index screen.
I would therefore suggest that the model gets modified, in order to allow more fields for every experiment entry. In particular, one can add "source language", "target language", "dataset name". This can be specified in the medatada file of the imported folder, or also be modified after the importing.
Additionally, the possibility of free tagging based on an additional sql table should be possible.
This fields can be used in a future implementation for sorting/filtering the main screen
The text was updated successfully, but these errors were encountered: