Skip to content

Update HTA dependency to the traceinsight package - #233

Merged
theodorbadea merged 8 commits into
mlcommons:mainfrom
Keysight:pr-217
Nov 5, 2026
Merged

theodorbadea merged 8 commits into
mlcommons:mainfrom
Keysight:pr-217

Conversation

@mihncalo

@mihncalo mihncalo commented Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

Summary

Reviewed the original PR from mlcommons/chakra ( #217 ) and addressed the comments left on that PR.

Test Plan

What was done

  • Pinned dependencies in pyproject.toml
  • Updated pyproject.toml to use requires-python 3.11 and ruff target-version py311, to prepare the environment ( these changes will come from the PR for Issue Raise requires-python to >=3.11 #224 )
  • Updated workflows to remove the HTA step
  • Updated USER_GUIDE.md

Environment used for testing

I used 2 environments - one before this PR and one with the changes in the current PR, to check that the HTA to traceinsight change and the introduced pins on dependencies do not affect the output and produce the same result.

I used 2 separate environments:

  • Old Environment - simulating the environment before introducing the changes in the current PR
  • New Environment - simulating the current environment, after introducing the changes in the current PR
    The purpose of the 2 environments was to check that they both produce the same results.

Old Environment:

  • Python 3.10.12
  • HTA ( the git pin - before replacing with traceinsight )
  • et_replay ( the git pin )
  • networkx 3.4.2
  • pandas 2.3.3
  • numpy 2.2.6

New Environment:

  • Python 3.13.17
  • traceinsight 0.6.1
  • et_replay ( the git pin )
  • networkx 3.7
  • pandas 3.1.1
  • numpy 2.5.4

How the testing was done

  • Ran python -m pytest tests: All 117 tests are passing, 1 warning: DeprecationWarning: Python 3.14 will, by default, filter extracted tar archives and reject files or modify their metadata. Use the filter argument to control this behavior

Input Files

  • single_rank - generated using trace_collection/post_execution/simple_onerank.py
  • multi_rank - borrowed traces from another project

Output Files

  • Obtained by running trace_collection/post_execution/combine_trace.sh
    • Results: linked traces, converted traces, jsonized files
  • Additionally, I ran the visualizer for graphml to compare the graphs

Results

The logs show that:

  • critical path runs: critical_path_analysis.py:1487 [INFO]: calculating critical path took 0.001714 seconds
  • the host and device traces are linked successfully: Linking process successful. Output file is available at ...
  • the linked traces are converted into .et files successfully: Conversion successful. Output file is available at ...

The outputs were identical

Closes #235

@github-actions

Copy link
Copy Markdown

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅

Comment thread pyproject.toml Resolved

@theodorbadea theodorbadea left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

PR description needs to be updated

Comment thread pyproject.toml Resolved
@theodorbadea

Copy link
Copy Markdown
Contributor

Jinsun also suggested for a DAG comparison in the original PR as a sanity check so that should be also done to be sure the graphs are ok.

@theodorbadea theodorbadea left a comment •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the concern with 3.6.0 is visible in the original description

Comment thread pyproject.toml Resolved
@theodorbadea
theodorbadea merged commit f5df2a1 into mlcommons:main Nov 5, 2026
10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Replace the git-pinned HTA dependency with traceinsight

3 participants