Logging learning rate #147
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
What does group number refer to? |
jon-tow
requested changes
Dec 22, 2022
| @@ -281,6 +281,8 @@ def learn(self): | |||
|
|
|||
| stats["time/forward"] = forward_time | |||
| stats["time/backward"] = backward_time | |||
| for group_number, lr in enumerate(self.scheduler.get_last_lr()): | |||
| stats[f"learning_rate/group_{group_number}"] = lr | |||
There was a problem hiding this comment.
Currently, we're logging a ton of metrics:
so to avoid adding extra clutter, I think it might be best to keep the LR dynamics tracked in Charts by using the following stats key instead: stats[f"learning_rate_group_{group_number}"] = lr
There was a problem hiding this comment.
Agree. Moved it to charts
An index number of a particular parameter group. In most cases there will be only one but we will also plot correctly in case of several groups |
jon-tow
approved these changes
Dec 23, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In this PR, learning rate is added to stats. Now it is easy to verify the LR scheduler work in wandb: