Skip to content
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.

Sign up for GitHub

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

Jump to bottom

capacitron training fixes #2086

Merged
merged 1 commit into from
Nov 1, 2022
Merged

capacitron training fixes #2086

merged 1 commit into from
Nov 1, 2022

Conversation

victor-shepardson
Copy link
Contributor

I needed these changes to get the recipes for capacitron working correctly

  • Trainer seems to expect CapacitronOptimizer to handle set_to_none argument
  • it appeared that when no special sampler is used, the training set was not being shuffled.

re: the shuffling, these sawtooth patterns appeared in both the step time and loss, but vanished when I set shuffle=True
Screenshot 2022-10-18 at 20 38 37

@CLAassistant
Copy link

CLAassistant commented Oct 18, 2022 •
edited
Loading

CLA assistant check
All committers have signed the CLA.

@erogol
Copy link
Member

erogol commented Oct 25, 2022

Thanks for the pr

@WeberJulian can you review it pls?

WeberJulian
WeberJulian approved these changes Oct 28, 2022
View reviewed changes
Copy link
Contributor

@WeberJulian WeberJulian left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me, thanks for the PR :)

@erogol erogol merged commit 5307a22 into coqui-ai:dev Nov 1, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@WeberJulian WeberJulian WeberJulian approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@victor-shepardson @CLAassistant @erogol @WeberJulian