tensorflow / tensorflow/tensorboard

Display final device placement

Open
#952 4 comments 4 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

Hi,

I was trying to use the "Color by device" option and my whole graph was gray. I learnt from this question that tensorboard only sees device assignments which are explicit, on the user side.

What would it take to show final device assignments, as they are when the graph is executed ? I realise this would have to be specified upfront, when actually saving the graph. Moreover, I believe that the main problem is that a graph is in no way bound to specific devices. This is what makes graphs and models shareable, in the first place.

But maybe the assignment could be written as a summary, much like all other stuff ?
Sorry if this issue is more on the tensorflow side than on the tensorboard, but it is unclear where implementation would have to start.

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 file, test, or entry point is named. Start by locating how the “Color by device” option receives device assignments and how graph summaries are saved; completion would mean showing final execution device assignments when that information was recorded, with behavior defined for graphs without it.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
data-visualization
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.