tensorflow / tensorflow/tensorboard

tensorboard line stops at some step

Open
#6,933 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

As this stackoverflow question poses: https://stackoverflow.com/questions/79124499/tensorboard-line-stops-at-some-step

image

The blue curve stuck at step 152 while other curves showed up. Any possible cause for this to happen?

I am using pytorch with tensor board. The code looks like this:

        tb_writer.add_scalars(
            main_tag="Train/{}CacheRates".format(name),
            tag_scalar_dict={
                "HR": hr,
                "#AdmitsRate": (
                    (
                        admit_action.sum(dim=1).float()+1
                    ) / (
                        mask.sum(dim=1).float()+1
                    )
                ).mean(),
                "AskedToPrefetchRate": (
                    (
                        is_prefech.sum(dim=1)+1
                    ).float() / (
                        num_items.float()+1
                    )
                ).mean(),
                "PrefetchGiveupRate": (
                    (
                        (is_prefech * (prefetch_action == 0).long()).sum(dim=1)+1
                    ).float() / (
                        is_prefech.sum(dim=1)+1
                    ).float()
                ).mean()
            },
            global_step=run_step,
        )

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

Start with the linked Stack Overflow question and reproduce the behavior using the shown PyTorch TensorBoard add_scalars call, especially around global_step 152. Compare the logged scalar series and TensorBoard display at that step; done means identifying a reproducible cause or documenting the missing information needed to diagnose it.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
data-visualization
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.