8000 Refactor: some fixes and new schedulers added by sarthakpati · Pull Request #33 · mlcommons/GaNDLF · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Refactor: some fixes and new schedulers added #33

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
Merged
merged 26 commits into from
Mar 30, 2021
Merged

Refactor: some fixes and new schedulers added #33

merged 26 commits into from
Mar 30, 2021

Conversation

sarthakpati
Copy link
Collaborator
@sarthakpati sarthakpati commented Mar 28, 2021 •
edited
Loading

Fixes #31

Proposed Changes

  • normalize_by_val fixed
  • added a new verbose parameter that can be used to print additional information
  • added cpu utilization
  • added coverage configuration and related information
  • msdnet added to tests
  • scheduler added and tests updated for all schedulers

Checklist

  • I have read the CONTRIBUTING guide
  • My PR is based from the current GANDLF master
  • Non-breaking change (would not break existing functionality): if changes breaks current code, please provide as many details as possible
  • Function/class source code documentation added/updated
  • Usage documentation has been updated, if appropriate
  • History has been updated, if appropriate
  • Tests added or modified to cover the changes
  • Integration tests passed locally by running pytest

@sarthakpati sarthakpati linked an issue Mar 28, 2021 that may be closed by this pull request
Issue in the "normalize_by_val" function #31
Closed
@sarthakpati sarthakpati changed the title Refactor Refactor: some fixes and new schedulers added Mar 28, 2021
@sarthakpati sarthakpati requested a review from Geeks-Sid March 28, 2021 12:28
Geeks-Sid
Geeks-Sid approved these changes Mar 28, 2021
View reviewed changes
GANDLF/training_loop.py
@@ -299,6 +303,11 @@ def validate_network(model, valid_dataloader, params):
average_epoch_valid_metric[metric] = total_epoch_valid_metric[metric] / len(
valid_dataloader
)

if params['scheduler'] == "reduce-on-plateau":
Copy link
Collaborator

Choose a reason for hiding this comment

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

exactly what i wanted to avoid

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

I understand - but how do we get around this, then? Completely remove all schedulers?

@Geeks-Sid Geeks-Sid merged commit a6eb610 into mlcommons:refactor Mar 30, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Geeks-Sid Geeks-Sid Geeks-Sid 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.

Issue in the "normalize_by_val" function
2 participants
@sarthakpati @Geeks-Sid
0