[None][fix] Update dataset identifier for cnn_dailymail to use namespaced repo in quantization scripts

[yufeiwu-nv]

Signed-off-by: yufeiwu-nv 230315618+yufeiwu-nv@users.noreply.github.com

Summary by CodeRabbit

• Bug Fixes
  • Improved compatibility with newer HuggingFace Hub versions for calibration dataset loading.

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

• PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

• PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

• Test cases are provided for new code paths (see test instructions)

• If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

• Any new dependencies have been scanned for license and vulnerabilities

• CODEOWNERS updated if ownership changes

• Documentation updated as needed

• Update tava architecture diagram if there is a significant design change in PR.

• The reviewers assigned automatically/manually are appropriate for the PR.

• Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR adds dataset id remapping logic to two quantization dataloader functions to address HuggingFace Hub compatibility. When "cnn_dailymail" is passed as the dataset identifier, both functions now remap it to the namespaced repository "ccdv/cnn_dailymail" before calling load_dataset.

Changes

HuggingFace dataset id compatibility

Layer / File(s)	Summary
Remap bare cnn_dailymail across calibration loaders tensorrt_llm/quantization/quantize_by_modelopt.py	get_calib_dataloader and get_nemo_calib_dataloader both conditionally remap "cnn_dailymail" to "ccdv/cnn_dailymail" before invoking load_dataset to maintain compatibility with newer HuggingFace Hub versions.

Possibly related PRs

• NVIDIA/TensorRT-LLM#14671: Adjusts how the cnn_dailymail dataset is resolved, complementing this PR's HuggingFace id remapping with local path handling changes.
• NVIDIA/TensorRT-LLM#13722: Also modifies get_calib_dataloader() and get_nemo_calib_dataloader() logic for cnn_dailymail-related dataset inputs via different matching rules.

Suggested labels

Community want to contribute

Suggested reviewers

• longlee0622
• ruodil

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

🚥 Pre-merge checks | ✅ 3 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is entirely empty except for template content and a sign-off. Required sections like Description and Test Coverage are not filled in, providing no explanation of the issue, solution, or test coverage.	Provide a Description section explaining the issue (why bare dataset ID no longer works) and the solution. Add a Test Coverage section documenting how the changes are tested.
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and specifically describes the main change: updating dataset identifier for cnn_dailymail to use namespaced repo in quantization scripts, matching the code changes.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

⚔️ Resolve merge conflicts

• Resolve merge conflict in branch bug

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@tensorrt_llm/quantization/quantize_by_modelopt.py`:
- Around line 529-532: The remap for the "cnn_dailymail" dataset should point to
the canonical Hugging Face repo "abisee/cnn_dailymail" (and optionally pin the
documented revision 3.0.0) rather than "ccdv/cnn_dailymail"; update the
conditional in quantize_by_modelopt.py that handles dataset_name_or_dir (the
branches checking for "cnn_dailymail") to set dataset_name_or_dir =
"abisee/cnn_dailymail" (or "abisee/cnn_dailymail@3.0.0" if you want to pin the
version), and apply the same fix to both occurrences mentioned in the review
(the blocks around where dataset_name_or_dir is reassigned). Ensure the logic
still preserves the original intent of replacing the bare "cnn_dailymail" id.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 155ed7b2-48c7-475e-9660-e3862b0e0db0

📥 Commits

Reviewing files that changed from the base of the PR and between e1212ad and b2a911f.

📒 Files selected for processing (1)

• tensorrt_llm/quantization/quantize_by_modelopt.py

[yufeiwu-nv]

/bot run

[tensorrt-cicd]

PR_Github #51957 [ run ] triggered by Bot. Commit: e6c013d Link to invocation

[tensorrt-cicd]

PR_Github #51957 [ run ] completed with state FAILURE. Commit: e6c013d
/LLM/main/L0_MergeRequest_PR pipeline #41308 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

• Please check the failed tests and fix your PR
• If you cannot view the failures, ask the CI triggerer to share details
• Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

[yufeiwu-nv]

/bot skip --comment "only test list modify"

[tensorrt-cicd]

PR_Github #51979 [ skip ] triggered by Bot. Commit: b3fa2d4 Link to invocation

[tensorrt-cicd]

PR_Github #51979 [ skip ] completed with state SUCCESS. Commit: b3fa2d4
Skipping testing for commit b3fa2d4

Link to invocation