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
Is your feature request related to a problem? Please describe.
With the current setup it is not possible to pass the batch size to Optuna as the dataloader objects are being created in a different pipeline, prior to the training/ optuna pipeline.
Therefore this particular hyperparameter cannot tuned with Optuna.
Describe the solution you'd like
Create dataloader objects within the instructor and return pure data tensors from the data processing pipeline.
Describe alternatives you've considered
Create dataloader objects in a separate node and call this node in each optuna trial.
Additional context
N/A
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
With the current setup it is not possible to pass the batch size to Optuna as the dataloader objects are being created in a different pipeline, prior to the training/ optuna pipeline.
Therefore this particular hyperparameter cannot tuned with Optuna.
Describe the solution you'd like
Create dataloader objects within the instructor and return pure data tensors from the data processing pipeline.
Describe alternatives you've considered
Create dataloader objects in a separate node and call this node in each optuna trial.
Additional context
N/A
The text was updated successfully, but these errors were encountered: