-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[M0-Dev] Callbacks support #17
Labels
Milestone
Comments
pab1s
added a commit
that referenced
this issue
Apr 22, 2024
pab1s
added a commit
that referenced
this issue
Apr 22, 2024
pab1s
added a commit
that referenced
this issue
Apr 22, 2024
pab1s
added a commit
that referenced
this issue
Apr 22, 2024
pab1s
added a commit
that referenced
this issue
Apr 22, 2024
Actually, this is basically finished. However, in a future it could be interesting to refactor schedulers or even metrics into callbacks. By now, I will leave it as it is. |
pab1s
added a commit
that referenced
this issue
May 28, 2024
pab1s
added a commit
that referenced
this issue
May 28, 2024
pab1s
added a commit
that referenced
this issue
May 28, 2024
pab1s
added a commit
that referenced
this issue
Jun 24, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I want to refactor the code to support callbacks. It is necessary to implement early stopping, so I decided that the best way to do it will be using callbacks. Moreover, I will change the way I log to do it using callbacks. My intention is to make the code cleaner and easier to understand.
With respect to early stopping, I will consider the following hyperparameters:
monitor
: to set if I want to monitor train or valid losses.patience
: how many epochs does the model update the weights with no improvements.delta
: min difference of losses between epochs that are considered an improvement.Probably I will also add support to verbosity, so it is easier to track.
The text was updated successfully, but these errors were encountered: