tensorflow / tensorflow/tensorboard

Option to indicate on a curve the epochs where learning rate changed

Open
#3,669 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

plugin:scalars type:feature
Dominant language
TypeScript
Stars
7.2k
Forks
1.7k
Avg merge
4d 22h
Merged PRs (30d)
1

Description

We often look both at our metric curve and the LR curve to see if an improvement on our metric value happened just after a change in the LR value.
It would be nice if we could have an option to automatically show the epochs at which the LR changed on any chosen curve.
"Indication" could be just a vertical line on the plot at each epoch where the learning rate changed.
It could be implemented with a checkbox under a curve to activate/deactivate LR change indications and a dropdown list to choose the name of the LR tag (so that it would work whatever the LR tag is).
This is particularly useful with ReduceLROnPlateau callback where the LR changes are rare.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No files, tests, or entry points are named. Start by locating the curve rendering and curve controls, then trace how learning-rate tags and epoch data are exposed; done means a selectable LR tag can mark its change epochs on a chosen curve with an on/off control.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
data-visualization, frontend
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.