tensorflow / tensorflow/tensorboard

tensorboard_profiling_keras.ipynb fails even in colab: logs -> $logs, No step marker found

Open
#6,954 1 comment 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

Environment information (required)

colab
https://github.com/tensorflow/tensorboard/blob/master/docs/tensorboard_profiling_keras.ipynb

Issue description

tensorboard profiler will only find profile information if pointed to the exact subdirectory where log is written ( ie single run)

therefore code needs to be changed to %tensorboard --logdir=$logs from %tensorboard --logdir=logs (to include the timestamp ie logs/ )

when this is done, profile tab is visible (and can select profile in the dropdown inactive/...

profile information is now visible, but on the overview page it warns

No step marker observed and hence the step time is unknown. This may happen if (1) training steps are not instrumented (e.g., if you are not using Keras) or (2) the profiling duration is shorter than the step time. For (1), you need to add step instrumentation; for (2), you may try to profile longer.

all timings are reported as zero on overview page
Screenshot 2024-11-24 at 10 58 09

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

Open docs/tensorboard_profiling_keras.ipynb and run it in Colab, starting with the %tensorboard --logdir cell and the profiling setup. Verify whether using $logs exposes the profile and whether the overview still reports no step marker and zero timings; done means the notebook's documented flow produces usable profile timings.

Written by the indexing model from the issue text.

Assessment

Tech stack
jupyter-notebook, keras
Domain
documentation, machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 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.