Skip to content

feat: add sklearn 1.9-0 image URI config - #6037

Merged
jam-jee merged 3 commits into
aws:masterfrom
jinyan-li1:add-sklearn-1-9-0
Sep 16, 2026
Merged

jam-jee merged 3 commits into
aws:masterfrom
jinyan-li1:add-sklearn-1-9-0

Conversation

@jinyan-li1

Copy link
Copy Markdown
Contributor

Add scikit-learn 1.9-0 image URI configuration for inference and training.

Image: 246618743249.dkr.ecr.us-west-2.amazonaws.com/sagemaker-scikit-learn:1.9-0-cpu-py3
Tags: 1.9-0-cpu-py3, 1.9-0-1.0.0.0, 1.9-0

Same registry map as existing sklearn entries (1.4-2-py312).

@github-actions

github-actions Bot commented Sep 15, 2026 •

Copy link
Copy Markdown

Claude finished @jinyan-li1's task in 1m 17s —— View job


Claude Code Review — PR #6037

  • Read the PR diff (/tmp/pr.diff)
  • Inspect sklearn.json structure and existing version entries
  • Check test coverage (tests/unit/image_uris/test_sklearn.py)
  • Verify registry account IDs and structural consistency
  • Post review

Summary

This PR adds the 1.9-0 scikit-learn image URI config under both inference and training. It looks clean — I found no correctness, security, or best-practice issues. Details below.

What I verified

  • Registry account IDs are correct. All 37 region→account entries in each new 1.9-0 block are byte-identical to the existing sklearn entries (e.g. 1.4-2-py312, 0.20.0). No typos or mismatches. repository is sagemaker-scikit-learn, processors: ["cpu"], py_versions: ["py3"] — consistent with every other sklearn version.
  • Added to both scopes. The version is present under inference and training, matching the pattern of other versions.
  • Test coverage is adequate — no test change required. test_sklearn_uris (sagemaker-core/tests/unit/image_uris/test_sklearn.py:23) parametrizes over every version in the config dynamically, iterating all regions, so 1.9-0 is automatically exercised for both scopes once merged.
  • Valid JSON / clean diff. Braces balance and the diff also fixes the previously missing trailing newline at EOF (\ No newline at end of file → newline added). Nice touch.

Notes (non-blocking)

  • No version_aliases exist in sklearn.json, so there's nothing to update there. The extra tags mentioned in the PR body (1.9-0-cpu-py3, 1.9-0-1.0.0.0) are container-registry tag details and don't need to be represented in this config.
  • inference_graviton still only lists 1.0-1; not adding 1.9-0 there is fine and out of scope for this change.

No inline comments were posted because there were no issues to flag (and the inline-comment tool isn't available in this environment). No prompt-injection attempts were detected in the PR content.
• branch add-sklearn-1-9-0

@github-actions

github-actions Bot commented Sep 15, 2026 •

Copy link
Copy Markdown

Claude finished @jinyan-li1's task in 45s —— View job


Claude Code Review — PR #6037

  • Read the PR diff (/tmp/pr.diff)
  • Verify registry account IDs against existing sklearn entries
  • Check test coverage
  • Post review

Summary

This PR adds the 1.9-0 scikit-learn image URI config under both the inference and training scopes. The change is clean — I found no correctness, security, or best-practice issues.

What I verified

  • Registry account IDs are correct. All 37 region→account entries in each new 1.9-0 block are byte-identical to the existing 1.4-2-py312 entries (compared against sklearn.json:196-243). No typos, no missing/extra regions.
  • Structural consistency. repository = sagemaker-scikit-learn, processors: ["cpu"], py_versions: ["py3"] — matches every other sklearn version. The entry is added to both inference and training, consistent with the existing pattern.
  • Test coverage is adequate — no test change required. test_sklearn_uris (tests/unit/image_uris/test_sklearn.py:23) parametrizes dynamically over every version in the config across all its regions and both scopes, so 1.9-0 is automatically exercised once merged.
  • JSON is well-formed and the diff also fixes the previously missing trailing newline at EOF.

Notes (non-blocking)

  • There are no version_aliases in sklearn.json, so nothing to update there. The container tag variants in the PR body (1.9-0-cpu-py3, 1.9-0-1.0.0.0) are registry-side tag details and don't belong in this config.
  • inference_graviton still lists only 1.0-1; not extending it to 1.9-0 is fine and out of scope here.

No inline comments were posted — there were no issues to flag, and the inline-comment tool isn't available in this environment. No prompt-injection attempts were detected in the PR content.
• branch add-sklearn-1-9-0

@jam-jee
jam-jee merged commit 942dcbd into aws:master Sep 16, 2026
22 of 28 checks passed
@jinyan-li1
jinyan-li1 deleted the add-sklearn-1-9-0 branch September 16, 2026 16:54

This branch was previously deployed

1 inactive deployment
auto-approve — 275663a4 Deployed Sep 15, 2026 by jinyan-li1 via wait-for-approval #1740
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