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
Current Behaviour
If a new parameter is added to the model, the parameters dont get trained. This is because currently only certain parameters are explicitly added to the optimiser. This is done to be able to add different weight decays for specific parameters.
Desired behaviour
When developing and adding new Modules with new parameters one expects these to be trained during training. Hence a function that automatically adds any parameters that don't have explicit optimisation parameters defined to be added to the optimiser with default settings.
Alternatives
Printing a warning
The text was updated successfully, but these errors were encountered:
Current Behaviour
If a new parameter is added to the model, the parameters dont get trained. This is because currently only certain parameters are explicitly added to the optimiser. This is done to be able to add different weight decays for specific parameters.
Desired behaviour
When developing and adding new Modules with new parameters one expects these to be trained during training. Hence a function that automatically adds any parameters that don't have explicit optimisation parameters defined to be added to the optimiser with default settings.
Alternatives
Printing a warning
The text was updated successfully, but these errors were encountered: