Skip to content

Fix typos and a stale model reference in the docs - #2744

Open
Oke Iyanuoluwa Enoch (Iyanuoluwa007) wants to merge 1 commit into
microsoft:mainfrom
Iyanuoluwa007:fix-docs-typos
Open

Oke Iyanuoluwa Enoch (Iyanuoluwa007) wants to merge 1 commit into
microsoft:mainfrom
Iyanuoluwa007:fix-docs-typos

Conversation

@Iyanuoluwa007

Copy link
Copy Markdown

Describe your changes

Fixes the documentation typos and inaccuracies listed in the codebase review (#2516), plus two more found with codespell:

  • README.md: "the the", "documentd".
  • notebooks/README.md: "ocmmand", "hot wo".
  • docs/source/how-to/index.md: the CLI command is capture-onnx-graph, not onnx-graph-capture. The list now puts optimize first instead of the deprecated auto-opt.
  • docs/source/getting-started/getting-started.md:
    • The text said the optimize example downloads Llama-3.2-1B-Instruct, but the command right below uses Qwen/Qwen2.5-0.5B-Instruct. The text now names the right model.
    • Fixed the grammar ("and optimizing the graph").
    • Dropped the fixed "around 60secs" estimate, since the run time depends on the hardware.
  • docs/source/features/onnx-transformations.md: "atleast".
  • docs/source/features/quantization.md: "simlar".

Docs only, no code changes. codespell reports nothing on README.md, notebooks/README.md and docs/source after this change.

Checklist before requesting a review

  • Add unit tests for this change.
  • Make sure all tests can pass.
  • Update documents if necessary.
  • Lint and apply fixes to your code by running lintrunner -a
  • Is this a user-facing change? If yes, give a description of this change to be included in the release notes.

(Optional) Issue link

Part of #2516

Copilot AI balanced review requested due to automatic review settings October 17, 2026 09:30
From the documentation items in the codebase review:

- README: "the the", "documentd".
- notebooks/README: "ocmmand", "hot wo".
- how-to index: the CLI command is capture-onnx-graph, not
  onnx-graph-capture; list optimize first instead of the deprecated
  auto-opt.
- getting started: the optimize example uses Qwen2.5-0.5B-Instruct, not
  Llama-3.2-1B-Instruct; fix the grammar and drop the fixed 60s estimate.
- Two more typos found with codespell ("atleast", "simlar").

Part of microsoft#2516
@azure-pipelines

Copy link
Copy Markdown
Azure Pipelines:
There may be pipelines that require an authorized user to comment /azp run to run.

Copilot AI 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.

🟢 Approval recommended

The documentation-only corrections match their context and leave no unresolved review findings.

No open issues 🎉

What changed in this PR

Corrects Olive documentation typos and outdated references, addressing part of #2516 without changing code.

Changes:

  • Fixes spelling errors across READMEs and feature documentation.
  • Corrects the CLI command list and prioritizes optimize.
  • Aligns the getting-started model description with its example and removes the fixed runtime estimate.
File Description
README.md Fixes spelling and duplicated wording.
notebooks/​README.md Corrects notebook link descriptions.
docs/​source/​how-to/​index.md Updates CLI command names and ordering.
docs/​source/​getting-started/​getting-started.md Corrects the model reference, grammar, and runtime guidance.
docs/​source/​features/​quantization.md Fixes spelling in the SpinQuant description.
docs/​source/​features/​onnx-transformations.md Corrects “at least” in transformation documentation.

Review effort: Balanced

This branch has not been deployed

No deployments
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.

2 participants