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
Once a model is trained, users have to inspect the checkpoints folder to know what has been saved. There is no indication of the loss or metrics corresponding to the validation set and the particular checkpoint.
The CAREamist should be able to print this information to users, it has access to the loss (and metrics in the future) for each epoch. Then it could also allow users to choose a checkpoint and load it, via maybe an epoch_number, rather than have them load manually via its path.
The text was updated successfully, but these errors were encountered:
Problem
Once a model is trained, users have to inspect the checkpoints folder to know what has been saved. There is no indication of the loss or metrics corresponding to the validation set and the particular checkpoint.
The
CAREamist
should be able to print this information to users, it has access to the loss (and metrics in the future) for each epoch. Then it could also allow users to choose a checkpoint and load it, via maybe anepoch_number
, rather than have them load manually via its path.The text was updated successfully, but these errors were encountered: