url string | repository_url string | labels_url string | comments_url string | events_url string | html_url string | id int64 | node_id string | number int64 | title string | user dict | labels list | state string | locked bool | assignees list | milestone dict | comments int64 | created_at timestamp[s] | updated_at timestamp[s] | closed_at timestamp[s] | assignee dict | author_association string | type dict | active_lock_reason string | sub_issues_summary dict | issue_dependencies_summary dict | body string | closed_by dict | reactions dict | timeline_url string | performed_via_github_app null | state_reason string | pinned_comment dict | draft bool | pull_request dict | is_pull_request bool |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33470 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33470/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33470/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33470/events | https://github.com/scikit-learn/scikit-learn/issues/33470 | 4,032,063,872 | I_kwDOAAzd1s7wVGmA | 33,470 | ⚠️ CI failed on Unit tests Linux x86-64 pylatest_free_threaded (last failure: Mar 06, 2026) ⚠️ | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Needs%20Triage",
"name": "Needs Triage",
"color": "FF0000",
"default": false,
"description": "Issue requires triage"
}
] | open | false | [] | null | 1 | 2026-03-06T03:52:08 | 2026-03-06T11:27:11 | null | null | CONTRIBUTOR | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | **CI failed on [Unit tests Linux x86-64 pylatest_free_threaded](https://github.com/scikit-learn/scikit-learn/actions/runs/22747548461/job/65974717148)** (Mar 06, 2026)
- test_estimators[RANSACRegressor(estimator=LinearRegression(),max_trials=10)-check_regressors_train]
- test_estimators[RANSACRegressor(estimator=LinearRegression(),max_trials=10)-check_regressors_train(readonly_memmap=True)]
- test_estimators[RANSACRegressor(estimator=LinearRegression(),max_trials=10)-check_regressors_train(readonly_memmap=True,X_dtype=float32)] | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33470/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33470/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33469 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33469/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33469/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33469/events | https://github.com/scikit-learn/scikit-learn/pull/33469 | 4,031,718,646 | PR_kwDOAAzd1s7IX8dM | 33,469 | [Callbacks] ENH Add fit-level callback support to Pipeline | {
"login": "gustavorubim",
"id": 7282895,
"node_id": "MDQ6VXNlcjcyODI4OTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/7282895?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gustavorubim",
"html_url": "https://github.com/gustavorubim",
"followers_url": "https://api.github.com/users/gustavorubim/followers",
"following_url": "https://api.github.com/users/gustavorubim/following{/other_user}",
"gists_url": "https://api.github.com/users/gustavorubim/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gustavorubim/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gustavorubim/subscriptions",
"organizations_url": "https://api.github.com/users/gustavorubim/orgs",
"repos_url": "https://api.github.com/users/gustavorubim/repos",
"events_url": "https://api.github.com/users/gustavorubim/events{/privacy}",
"received_events_url": "https://api.github.com/users/gustavorubim/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815450942,
"node_id": "MDU6TGFiZWwxODE1NDUwOTQy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:pipeline",
"name": "module:pipeline",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-03-06T01:51:41 | 2026-03-06T10:46:57 | 2026-03-06T10:46:57 | null | NONE | null | null | null | null | #### Reference Issues/PRs
Towards #27676
#### What does this implement/fix? Explain your changes.
- add callback support to `Pipeline.fit` by emitting one callback task per pipeline step
- keep the `Pipeline` step callbacks observational only: callback stop values are ignored
- support step callbacks for regular, `None` / `"passthrough"`, and cached steps
- add tests covering step payloads, passthrough steps, and `memory=` cache hits
#### Any other comments?
This is an intentionally narrow first `Pipeline` integration. It covers fit-level step callbacks only and does not propagate callbacks into sub-estimators.
This pull request includes code written with the assistance of AI.
The code has been reviewed by a human. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33469/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33469/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33469",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33469",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33469.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33469.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33468 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33468/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33468/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33468/events | https://github.com/scikit-learn/scikit-learn/pull/33468 | 4,031,562,842 | PR_kwDOAAzd1s7IXcDA | 33,468 | [Callbacks] ENH Add callback support to KMeans and MiniBatchKMeans | {
"login": "gustavorubim",
"id": 7282895,
"node_id": "MDQ6VXNlcjcyODI4OTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/7282895?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gustavorubim",
"html_url": "https://github.com/gustavorubim",
"followers_url": "https://api.github.com/users/gustavorubim/followers",
"following_url": "https://api.github.com/users/gustavorubim/following{/other_user}",
"gists_url": "https://api.github.com/users/gustavorubim/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gustavorubim/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gustavorubim/subscriptions",
"organizations_url": "https://api.github.com/users/gustavorubim/orgs",
"repos_url": "https://api.github.com/users/gustavorubim/repos",
"events_url": "https://api.github.com/users/gustavorubim/events{/privacy}",
"received_events_url": "https://api.github.com/users/gustavorubim/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815435821,
"node_id": "MDU6TGFiZWwxODE1NDM1ODIx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:cluster",
"name": "module:cluster",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-03-06T01:06:33 | 2026-03-06T10:46:52 | 2026-03-06T10:46:51 | null | NONE | null | null | null | null | #### Reference Issues/PRs
Towards #27676
#### What does this implement/fix? Explain your changes.
- add callback support to `KMeans` by emitting nested `init` and `iter` tasks during `fit`
- honor callback stop signals at both the iteration and initialization levels in `KMeans`
- add callback support to `MiniBatchKMeans` by emitting one callback task per minibatch step during `fit`
- honor callback stop signals to terminate `MiniBatchKMeans.fit` early
- emit a callback task from `MiniBatchKMeans.partial_fit` for consistency with `fit`
- add tests covering nested hook calls and callback-driven stopping
#### Any other comments?
This is intended as a first production estimator integration on top of the `callbacks` branch.
This pull request includes code written with the assistance of AI.
The code has been reviewed by a human. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33468/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33468/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33468",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33468",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33468.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33468.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33467 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33467/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33467/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33467/events | https://github.com/scikit-learn/scikit-learn/pull/33467 | 4,028,265,788 | PR_kwDOAAzd1s7IM8mO | 33,467 | DOC mention Newton boosting for HGBT in user guide | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 3730476391,
"node_id": "LA_kwDOAAzd1s7eW... | open | false | [] | null | 0 | 2026-03-05T13:34:22 | 2026-03-06T02:47:35 | null | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
None
#### What does this implement/fix? Explain your changes.
#### AI usage disclosure
None
#### Any other comments?
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33467/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33467/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33467",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33467",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33467.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33467.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33466 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33466/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33466/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33466/events | https://github.com/scikit-learn/scikit-learn/pull/33466 | 4,028,211,277 | PR_kwDOAAzd1s7IMxJ6 | 33,466 | DOC better marketing in user guid for Array API and set_output pandas/polars | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
}
] | open | false | [] | null | 0 | 2026-03-05T13:23:32 | 2026-03-05T13:23:58 | null | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
None
#### What does this implement/fix? Explain your changes.
The current user guide makes it hard to find "array API" or "pandas/polars output transform" unless one searches for it.
#### AI usage disclosure
None
#### Any other comments? | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33466/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33466/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33466",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33466",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33466.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33466.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33465 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33465/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33465/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33465/events | https://github.com/scikit-learn/scikit-learn/pull/33465 | 4,026,835,243 | PR_kwDOAAzd1s7IILit | 33,465 | TST use global_random_seed in sklearn/linear_model/tests/test_bayes.py | {
"login": "gitbalaji",
"id": 1065555,
"node_id": "MDQ6VXNlcjEwNjU1NTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/1065555?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gitbalaji",
"html_url": "https://github.com/gitbalaji",
"followers_url": "https://api.github.com/users/gitbalaji/followers",
"following_url": "https://api.github.com/users/gitbalaji/following{/other_user}",
"gists_url": "https://api.github.com/users/gitbalaji/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gitbalaji/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gitbalaji/subscriptions",
"organizations_url": "https://api.github.com/users/gitbalaji/orgs",
"repos_url": "https://api.github.com/users/gitbalaji/repos",
"events_url": "https://api.github.com/users/gitbalaji/events{/privacy}",
"received_events_url": "https://api.github.com/users/gitbalaji/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
}
] | open | false | [] | null | 0 | 2026-03-05T08:56:40 | 2026-03-05T08:56:56 | null | null | NONE | null | null | null | null | Convert remaining tests with hardcoded random seeds to use the `global_random_seed` fixture, making them seed-insensitive.
**Converted tests:**
- `test_prediction_bayesian_ridge_ard_with_constant_input`: `check_random_state(42)` → `check_random_state(global_random_seed)`
- `test_std_bayesian_ridge_ard_with_constant_input`: `check_random_state(42)` → `check_random_state(global_random_seed)`
- `test_return_std`: bare `np.random.random()`/`np.random.randn()` → `rng = np.random.RandomState(global_random_seed)`
**Verification:**
- All 400 runs (4 test variants × 100 seeds) pass with `SKLEARN_TESTS_GLOBAL_RANDOM_SEED="all"`
Ref #22827 | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33465/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33465/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33465",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33465",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33465.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33465.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33463 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33463/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33463/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33463/events | https://github.com/scikit-learn/scikit-learn/pull/33463 | 4,025,690,948 | PR_kwDOAAzd1s7IEWBl | 33,463 | Added contribution file | {
"login": "vedpathakpranav",
"id": 182463023,
"node_id": "U_kgDOCuAqLw",
"avatar_url": "https://avatars.githubusercontent.com/u/182463023?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vedpathakpranav",
"html_url": "https://github.com/vedpathakpranav",
"followers_url": "https://api.github.com/users/vedpathakpranav/followers",
"following_url": "https://api.github.com/users/vedpathakpranav/following{/other_user}",
"gists_url": "https://api.github.com/users/vedpathakpranav/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vedpathakpranav/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vedpathakpranav/subscriptions",
"organizations_url": "https://api.github.com/users/vedpathakpranav/orgs",
"repos_url": "https://api.github.com/users/vedpathakpranav/repos",
"events_url": "https://api.github.com/users/vedpathakpranav/events{/privacy}",
"received_events_url": "https://api.github.com/users/vedpathakpranav/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2399338178,
"node_id": "MDU6TGFiZWwyMzk5MzM4MTc4",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/spam",
"name": "spam",
"color": "b60205",
"default": false,
"description": "spam PR and issues"
}
] | closed | false | [] | null | 0 | 2026-03-05T04:34:36 | 2026-03-05T05:01:27 | 2026-03-05T05:01:24 | null | NONE | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33463/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33463/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33463",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33463",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33463.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33463.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33459 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33459/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33459/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33459/events | https://github.com/scikit-learn/scikit-learn/pull/33459 | 4,025,042,837 | PR_kwDOAAzd1s7ICOSU | 33,459 | FIX account for sample_weight in SAGA step size computation | {
"login": "ahmvdd",
"id": 112165741,
"node_id": "U_kgDOBq-DbQ",
"avatar_url": "https://avatars.githubusercontent.com/u/112165741?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ahmvdd",
"html_url": "https://github.com/ahmvdd",
"followers_url": "https://api.github.com/users/ahmvdd/followers",
"following_url": "https://api.github.com/users/ahmvdd/following{/other_user}",
"gists_url": "https://api.github.com/users/ahmvdd/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ahmvdd/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ahmvdd/subscriptions",
"organizations_url": "https://api.github.com/users/ahmvdd/orgs",
"repos_url": "https://api.github.com/users/ahmvdd/repos",
"events_url": "https://api.github.com/users/ahmvdd/events{/privacy}",
"received_events_url": "https://api.github.com/users/ahmvdd/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-03-05T01:05:04 | 2026-03-05T14:25:45 | 2026-03-05T08:44:49 | null | NONE | null | null | null | null | The step size for SAGA was computed using max(||x_i||^2) without accounting for sample weights. When sample_weight[i] > 1, the effective Lipschitz constant of sample i is w_i * L_i, so the global L must use max(w_i * ||x_i||^2) as an upper bound.
Add a regression test comparing SAGA with non-uniform sample_weight against lbfgs as a reference solver.
<!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33459/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33459/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33459",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33459",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33459.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33459.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33457 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33457/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33457/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33457/events | https://github.com/scikit-learn/scikit-learn/issues/33457 | 4,024,081,978 | I_kwDOAAzd1s7v2p46 | 33,457 | META track `sample_weight` support for all estimators | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3941176996,
"node_id": "LA_kwDOAAzd1s7q6Zak",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Meta-issue",
"name": "Meta-issue",
"color": "B645AE",
"default": false,
"description": "General issue associated to an identified list of tasks"
}
] | open | false | [] | null | 0 | 2026-03-04T20:50:29 | 2026-03-05T08:00:44 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | The goal of this issue is to track the support of `estimator.fit(X, y, sample_weight=..)` throughout the library. As of scikit-learn version 1.8:
- [ ] `sklearn.cluster.AffinityPropagation`
- [ ] `sklearn.cluster.AgglomerativeClustering`
- [ ] `sklearn.cluster.Birch`
- [ ] `sklearn.cluster.FeatureAgglomeration`
- [ ] `sklearn.cluster.HDBSCAN`
- [ ] `sklearn.cluster.MeanShift`
- [ ] `sklearn.cluster.OPTICS`
- [ ] `sklearn.cluster.SpectralClustering`
- [ ] `sklearn.compose.TransformedTargetRegressor`
- [ ] `sklearn.cross_decomposition.CCA`
- [ ] `sklearn.cross_decomposition.PLSCanonical`
- [ ] `sklearn.cross_decomposition.PLSRegression`
- [ ] `sklearn.discriminant_analysis.LinearDiscriminantAnalysis`
- [ ] `sklearn.discriminant_analysis.QuadraticDiscriminantAnalysis`
- [x] `sklearn.ensemble.HistGradientBoostingClassifier` and `HistGradientBoostingRegressor` binning (added manually)
#29641
- [ ] `sklearn.ensemble.StackingClassifier`
- [ ] `sklearn.ensemble.StackingRegressor`
- [ ] `sklearn.ensemble.VotingClassifier`
- [ ] `sklearn.ensemble.VotingRegressor`
- [ ] `sklearn.gaussian_process.GaussianProcessClassifier`
- [ ] `sklearn.gaussian_process.GaussianProcessRegressor`
- [ ] `sklearn.linear_model.ARDRegression`
- [ ] `sklearn.linear_model.Lars`
- [ ] `sklearn.linear_model.LarsCV`
- [ ] `sklearn.linear_model.LassoLars`
- [ ] `sklearn.linear_model.LassoLarsCV`
- [ ] `sklearn.linear_model.LassoLarsIC`
- [ ] `sklearn.linear_model.MultiTaskElasticNet`
#33440
- [ ] `sklearn.linear_model.MultiTaskElasticNetCV`
#33440
- [ ] `sklearn.linear_model.MultiTaskLasso`
#33440
- [ ] `sklearn.linear_model.MultiTaskLassoCV`
#33440
- [ ] `sklearn.linear_model.OrthogonalMatchingPursuit`
- [ ] `sklearn.linear_model.OrthogonalMatchingPursuitCV`
- [ ] `sklearn.linear_model.PassiveAggressiveClassifier`
deprecated
- [ ] `sklearn.linear_model.PassiveAggressiveRegressor`
deprecated
- [ ] `sklearn.linear_model.TheilSenRegressor`
- [ ] `sklearn.model_selection.FixedThresholdClassifier`
- [ ] `sklearn.model_selection.TunedThresholdClassifierCV`
- [ ] `sklearn.multiclass.OneVsOneClassifier`
- [ ] `sklearn.multiclass.OneVsRestClassifier`
- [ ] `sklearn.multioutput.ClassifierChain`
- [ ] `sklearn.multiclass.OutputCodeClassifier`
- [ ] `sklearn.multioutput.RegressorChain`
- [ ] `sklearn.neighbors.KNeighborsClassifier`
- [ ] `sklearn.neighbors.KNeighborsRegressor`
- [ ] `sklearn.neighbors.NearestCentroid`
- [ ] `sklearn.neighbors.RadiusNeighborsClassifier`
- [ ] `sklearn.neighbors.RadiusNeighborsRegressor`
- [ ] `sklearn.semi_supervised.LabelPropagation`
- [ ] `sklearn.semi_supervised.LabelSpreading`
- [ ] `sklearn.semi_supervised.SelfTrainingClassifier`
Code to generate that list
```py
from inspect import signature
import sklearn
from sklearn.utils import all_estimators
from sklearn.utils._test_common.instance_generator import _construct_instances
[est for name, est in all_estimators(type_filter=["classifier", "regressor", "cluster"]) if "sample_weig
⋮ ht" not in signature(next(_construct_instances(est)).fit).parameters]
``` | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33457/reactions",
"total_count": 2,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33457/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33456 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33456/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33456/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33456/events | https://github.com/scikit-learn/scikit-learn/pull/33456 | 4,022,865,533 | PR_kwDOAAzd1s7H7C5w | 33,456 | FEAT Add RollingTimeSeriesSplit class and tests | {
"login": "Dat-Al",
"id": 126704838,
"node_id": "U_kgDOB41cxg",
"avatar_url": "https://avatars.githubusercontent.com/u/126704838?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Dat-Al",
"html_url": "https://github.com/Dat-Al",
"followers_url": "https://api.github.com/users/Dat-Al/followers",
"following_url": "https://api.github.com/users/Dat-Al/following{/other_user}",
"gists_url": "https://api.github.com/users/Dat-Al/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Dat-Al/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Dat-Al/subscriptions",
"organizations_url": "https://api.github.com/users/Dat-Al/orgs",
"repos_url": "https://api.github.com/users/Dat-Al/repos",
"events_url": "https://api.github.com/users/Dat-Al/events{/privacy}",
"received_events_url": "https://api.github.com/users/Dat-Al/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815446740,
"node_id": "MDU6TGFiZWwxODE1NDQ2NzQw",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:model_selection",
"name": "module:model_selection",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 9691732262,
"node_id":... | closed | false | [] | null | 2 | 2026-03-04T16:20:45 | 2026-03-05T05:06:48 | 2026-03-05T05:06:48 | null | NONE | null | null | null | null | ### **Reference Issues/PRs**
Addresses a common need for fixed-window cross-validation (rolling origin) in time series, distinct from the expanding window strategy of `TimeSeriesSplit`.
### **What does this implement/fix? Explain your changes.**
This PR introduces a new cross-validator: `RollingTimeSeriesSplit`.
### **Motivation:**
The current `TimeSeriesSplit` implements an expanding window strategy (the training set grows over time). However, in many time-series applications (finance, physics, demand forecasting), it is standard to use a **fixed sliding window** (Walk-Forward Validation) to detect concept drift and validate model performance on strictly recent history.
While `TimeSeriesSplit` accepts a `max_train_size`, forcing it to behave like a strict rolling window with a specific step size is currently unintuitive and requires complex calculations of n_splits based on the dataset length.
### **Proposed Solution:**
`RollingTimeSeriesSplit` allows defining:
- `train_size` (fixed)
- `test_size` (fixed)
- `step_size` (optional, defaults to train_size for disjoint windows)
- `gap` (optional, for embargo/purge)
### **Key Feature - Right Alignment:**
Unlike standard implementations that might drop recent data if the division isn't perfect, this implementation is **right-aligned**. It anchors the splits to the end of the dataset (most recent data). Any historical "remainder" (indices at the start that don't fit a full fold) is added to the first training set. This ensures the model is always validated on the latest available data.
### **AI usage disclosure**
I used AI assistance for:
- [x] Code generation (e.g., when writing an implementation or fixing a bug)
- [x] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
### **Any other comments?**
I have added unit tests in `sklearn/model_selection/tests/test_split.py` covering:
- Disjoint windows (`step_size` = `train_size`)
- Overlapping windows (`step_size` < `train_size`)
- The `gap` parameter
- The right-alignment logic (remainder handling)
### **Example usage:**
```
import numpy as np
from sklearn.model_selection import RollingTimeSeriesSplit
X = np.arange(14)
# Scenario: Train on 4 samples, Test on 1 sample.
# Default step=4 (disjoint).
# Remainder (1 sample) goes to the first train set.
cv = RollingTimeSeriesSplit(train_size=4, test_size=1)
for i, (train, test) in enumerate(cv.split(X)):
print(f"Split {i}: Train={train}, Test={test}")
# Output:
# Split 0: Train=[0 1 2 3 4], Test=[5] <-- Absorbs index 0 (remainder)
# Split 1: Train=[5 6 7 8], Test=[9]
# Split 2: Train=[9 10 11 12], Test=[13] <-- Perfectly aligned with end
```
| {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33456/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33456/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33456",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33456",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33456.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33456.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33455 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33455/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33455/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33455/events | https://github.com/scikit-learn/scikit-learn/pull/33455 | 4,022,533,631 | PR_kwDOAAzd1s7H59zO | 33,455 | FIX: HDBSCAN TypeError with cluster_selection_epsilon on tied distances | {
"login": "repeating",
"id": 37001921,
"node_id": "MDQ6VXNlcjM3MDAxOTIx",
"avatar_url": "https://avatars.githubusercontent.com/u/37001921?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/repeating",
"html_url": "https://github.com/repeating",
"followers_url": "https://api.github.com/users/repeating/followers",
"following_url": "https://api.github.com/users/repeating/following{/other_user}",
"gists_url": "https://api.github.com/users/repeating/gists{/gist_id}",
"starred_url": "https://api.github.com/users/repeating/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/repeating/subscriptions",
"organizations_url": "https://api.github.com/users/repeating/orgs",
"repos_url": "https://api.github.com/users/repeating/repos",
"events_url": "https://api.github.com/users/repeating/events{/privacy}",
"received_events_url": "https://api.github.com/users/repeating/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815435821,
"node_id": "MDU6TGFiZWwxODE1NDM1ODIx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:cluster",
"name": "module:cluster",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2911057044,
"node_id": "MDU6TGFiZWwyOT... | closed | false | [] | null | 0 | 2026-03-04T15:15:57 | 2026-03-05T08:20:37 | 2026-03-05T08:20:37 | null | NONE | null | null | null | null | closes #33219
the `traverse_upwards` function in `_tree.pyx` does boolean-indexed lookups on the condensed tree structured array:
```python
parent = cluster_tree[cluster_tree['child'] == leaf]['parent']
parent_eps = 1 / cluster_tree[cluster_tree['child'] == parent]['value']
```
these always return shape-(1,) arrays, not scalars. numpy < 2.0 silently coerced shape-(1,) to scalar during cython typed assignment, but numpy 2.0 made this strict and raises `TypeError: only 0-dimensional arrays can be converted to Python scalars`.
the fix adds `[0]` to both lookups to explicitly extract the scalar. this is always safe since each child node has exactly one parent in the condensed tree.
added a non-regression test using the exact distance matrix from the issue.
## AI usage disclosure
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [x] Test/benchmark generation
- [ ] Documentation (including examples)
- [x] Research and understanding | {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33455/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33455/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33455",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33455",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33455.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33455.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33454 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33454/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33454/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33454/events | https://github.com/scikit-learn/scikit-learn/issues/33454 | 4,022,486,086 | I_kwDOAAzd1s7vwkRG | 33,454 | DictionaryLearner dies with 'omp' only | {
"login": "kleingeo",
"id": 31522137,
"node_id": "MDQ6VXNlcjMxNTIyMTM3",
"avatar_url": "https://avatars.githubusercontent.com/u/31522137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kleingeo",
"html_url": "https://github.com/kleingeo",
"followers_url": "https://api.github.com/users/kleingeo/followers",
"following_url": "https://api.github.com/users/kleingeo/following{/other_user}",
"gists_url": "https://api.github.com/users/kleingeo/gists{/gist_id}",
"starred_url": "https://api.github.com/users/kleingeo/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/kleingeo/subscriptions",
"organizations_url": "https://api.github.com/users/kleingeo/orgs",
"repos_url": "https://api.github.com/users/kleingeo/repos",
"events_url": "https://api.github.com/users/kleingeo/events{/privacy}",
"received_events_url": "https://api.github.com/users/kleingeo/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995755,
"node_id": "MDU6TGFiZWwxMjk5NTc1NQ==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Bug",
"name": "Bug",
"color": "e10c02",
"default": false,
"description": null
},
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "ht... | open | false | [] | null | 0 | 2026-03-04T15:06:09 | 2026-03-05T15:30:26 | null | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the bug and give evidence about its user-facing impact
I noticed new behaviour in `sklearn.decomposition.DictioanryLearning`. I have noticed an error that I have not seen before. Learner `transform_n_nonzero_coefs` > `n_components` and only when `transform_algorithm = "omp"`.
Example below is similar to my current use-case. Situation occurs due to large hyperparameter grid search.
Using newest version, `sklearn==1.8.0`.
Documentation should be updated to reflect the difference in outputs and limitations.
### Steps/Code to Reproduce
```python
import numpy as np
import sklearn
from sklearn.preprocessing import StandardScaler, OneHotEncoder, MinMaxScaler
from sklearn.compose import ColumnTransformer
if __name__ == '__main__':
column_scaler = ColumnTransformer(
transformers=[
('vert_encoder', OneHotEncoder(), [6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18]),
],
remainder='passthrough',
sparse_threshold=0,
)
N = 150
x_main = np.zeros((N, 26))
x_main[:, :6] = np.random.rand(N, 6)
x_main[:, 6] = np.random.randint(1, 25, size=N)
x_main[:, 7] = np.random.randint(1, 3, size=N)
x_main[:, 8] = np.random.randint(0, 5, size=N)
x_main[:, 9] = np.random.randint(1, 17, size=N)
x_main[:, 10] = np.random.randint(1, 3, size=N)
x_main[:, 11] = np.random.randint(1, 3, size=N)
x_main[:, 12] = np.random.randint(1, 3, size=N)
x_main[:, 13] = np.random.randint(0, 7, size=N)
x_main[:, 14] = np.random.randint(1, 4, size=N)
x_main[:, 15] = np.random.randint(0, 4, size=N)
x_main[:, 16] = np.random.randint(0, 4, size=N)
x_main[:, 17] = np.random.randint(0, 4, size=N)
x_main[:, 18] = np.random.randint(0, 19, size=N)
x_main[:, 19:] = np.random.rand(N, 7)
column_scaler.fit(x_main)
x_scaled = column_scaler.transform(x_main)
x_val_scaled = column_scaler.transform(x_main)
dict_params = {'n_components': 18,
'max_iter': 1000,
'transform_algorithm': 'lars',
'transform_n_nonzero_coefs': 19,
}
dl_lars = sklearn.decomposition.DictionaryLearning(random_state=0,
**dict_params)
dl_lars.fit(x_scaled)
x_scaled_transformed = dl_lars.transform(x_scaled)
print('passed (fake) training data')
x_val_scaled_transformed = dl_lars.transform(x_val_scaled)
print('passed (fake) validation data')
print('Changing to "omp"')
dict_params = {'n_components': 18,
'max_iter': 1000,
'transform_algorithm': 'omp',
'transform_n_nonzero_coefs': 19,
}
dl_omp = sklearn.decomposition.DictionaryLearning(random_state=0,
**dict_params)
dl_omp.fit(x_scaled)
x_scaled_transformed = dl_omp.transform(x_scaled)
print('passed (fake) training data')
x_val_scaled_transformed = dl_omp.transform(x_val_scaled)
print('passed (fake) validation data')
```
### Expected Results
No errors.
### Actual Results
```python
Connected to pydev debugger (build 253.28294.336)
passed (fake) training data
passed (fake) validation data
Changing to "omp"
Traceback (most recent call last):
File "C:\Program Files\JetBrains\PyCharm\plugins\python-ce\helpers\pydev\pydevd.py", line 1647, in _exec
pydev_imports.execfile(file, globals, locals) # execute the script
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Program Files\JetBrains\PyCharm\plugins\python-ce\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "C:\SpineFracturePrediction\test_sklearn_dl.py", line 69, in <module>
x_scaled_transformed = dl_omp.transform(x_scaled)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\utils\_set_output.py", line 316, in wrapped
data_to_wrap = f(self, X, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\decomposition\_dict_learning.py", line 1141, in transform
return self._transform(X, self.components_)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\decomposition\_dict_learning.py", line 1102, in _transform
code = sparse_encode(
^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\utils\_param_validation.py", line 218, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\decomposition\_dict_learning.py", line 374, in sparse_encode
return _sparse_encode(
^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\decomposition\_dict_learning.py", line 429, in _sparse_encode
code = _sparse_encode_precomputed(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\decomposition\_dict_learning.py", line 192, in _sparse_encode_precomputed
new_code = orthogonal_mp_gram(
^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\utils\_param_validation.py", line 191, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Miniforge3\envs\spine_frac_312\Lib\site-packages\sklearn\linear_model\_omp.py", line 605, in orthogonal_mp_gram
raise ValueError(
ValueError: The number of atoms cannot be more than the number of features
```
### Versions
```shell
sklearn.show_versions()
System:
python: 3.12.12 | packaged by conda-forge | (main, Jan 26 2026, 23:38:32) [MSC v.1944 64 bit (AMD64)]
executable: C:\Miniforge3\envs\spine_frac_312\python.exe
machine: Windows-11-10.0.26100-SP0
Python dependencies:
sklearn: 1.8.0
pip: 26.0.1
setuptools: 82.0.0
numpy: 2.4.2
scipy: 1.17.1
Cython: None
pandas: 3.0.1
matplotlib: 3.10.8
joblib: 1.5.3
threadpoolctl: 3.6.0
Built with OpenMP: True
threadpoolctl info:
user_api: blas
internal_api: mkl
num_threads: 8
prefix: libblas
filepath: C:\Miniforge3\envs\spine_frac_312\Library\bin\libblas.dll
version: 2020.0.4
threading_layer: intel
user_api: openmp
internal_api: openmp
num_threads: 16
prefix: libiomp
filepath: C:\Miniforge3\envs\spine_frac_312\Library\bin\libiomp5md.dll
version: None
user_api: openmp
internal_api: openmp
num_threads: 16
prefix: vcomp
filepath: C:\Miniforge3\envs\spine_frac_312\vcomp140.dll
version: None
```
### Interest in fixing the bug
No | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33454/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33454/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33453 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33453/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33453/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33453/events | https://github.com/scikit-learn/scikit-learn/pull/33453 | 4,022,257,941 | PR_kwDOAAzd1s7H5D4M | 33,453 | MNT Deprecate `shuffle` and `random_state` params from `TargetEncoder` | {
"login": "StefanieSenger",
"id": 91849487,
"node_id": "U_kgDOBXmDDw",
"avatar_url": "https://avatars.githubusercontent.com/u/91849487?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/StefanieSenger",
"html_url": "https://github.com/StefanieSenger",
"followers_url": "https://api.github.com/users/StefanieSenger/followers",
"following_url": "https://api.github.com/users/StefanieSenger/following{/other_user}",
"gists_url": "https://api.github.com/users/StefanieSenger/gists{/gist_id}",
"starred_url": "https://api.github.com/users/StefanieSenger/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/StefanieSenger/subscriptions",
"organizations_url": "https://api.github.com/users/StefanieSenger/orgs",
"repos_url": "https://api.github.com/users/StefanieSenger/repos",
"events_url": "https://api.github.com/users/StefanieSenger/events{/privacy}",
"received_events_url": "https://api.github.com/users/StefanieSenger/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815448196,
"node_id": "MDU6TGFiZWwxODE1NDQ4MTk2",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:preprocessing",
"name": "module:preprocessing",
"color": "151e91",
"default": false,
"description": ""
}
] | open | false | [] | null | 2 | 2026-03-04T14:20:25 | 2026-03-06T12:23:31 | null | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
https://github.com/scikit-learn/scikit-learn/pull/33089#discussion_r2704525722
#### What does this implement/fix? Explain your changes.
This PR deprecates the `shuffle` and `random_state` params from `TargetEncoder`. They are now redundant and had only been there in the first place because before #33089 only ints could be passed as `cv` arguments.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [x] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33453/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33453/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33453",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33453",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33453.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33453.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33451 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33451/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33451/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33451/events | https://github.com/scikit-learn/scikit-learn/pull/33451 | 4,020,962,415 | PR_kwDOAAzd1s7H0y-n | 33,451 | FIX check_array crash with pandas StringDtype as dtype argument | {
"login": "repeating",
"id": 37001921,
"node_id": "MDQ6VXNlcjM3MDAxOTIx",
"avatar_url": "https://avatars.githubusercontent.com/u/37001921?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/repeating",
"html_url": "https://github.com/repeating",
"followers_url": "https://api.github.com/users/repeating/followers",
"following_url": "https://api.github.com/users/repeating/following{/other_user}",
"gists_url": "https://api.github.com/users/repeating/gists{/gist_id}",
"starred_url": "https://api.github.com/users/repeating/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/repeating/subscriptions",
"organizations_url": "https://api.github.com/users/repeating/orgs",
"repos_url": "https://api.github.com/users/repeating/repos",
"events_url": "https://api.github.com/users/repeating/events{/privacy}",
"received_events_url": "https://api.github.com/users/repeating/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815453311,
"node_id": "MDU6TGFiZWwxODE1NDUzMzEx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:utils",
"name": "module:utils",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-03-04T09:20:55 | 2026-03-05T08:23:23 | 2026-03-05T08:23:23 | null | NONE | null | null | null | null | #### Reference Issues/PRs
Fixes #33383
#### What does this implement/fix? Explain your changes.
`check_array` has a block that normalizes the `dtype` argument by calling `np.dtype(dtype)` (line ~952 in `validation.py`). This works fine for numpy dtypes and string aliases like `"float32"`, but blows up with a `TypeError` for any pandas extension dtype that has no numpy equivalent — `pd.StringDtype` being the one that hits users in practice because it became the default dtype for string columns in pandas 3.
The other nullable extension dtypes (`Int64Dtype`, `Float64Dtype`, `BooleanDtype`) are handled earlier by `_pandas_dtype_needs_early_conversion()`, which converts the array upfront and sets `dtype=None` before reaching that line. `StringDtype` isn't handled there (it's not numeric/bool), so it falls through and crashes.
Fix is a small `try/except TypeError` around the `np.dtype()` call. If the dtype can't be represented in numpy, we reset it to `None` and let numpy pick the natural dtype for the array — which is `object` for string data, exactly what you'd get from `np.asarray(pd_string_series)` anyway. No data loss, fully backward-compatible, and it also handles `CategoricalDtype` and any future pandas extension dtype that hits the same path.
**Changed files:**
- `sklearn/utils/validation.py` — wrapped `np.dtype(dtype)` in try/except TypeError (6 lines)
- `sklearn/utils/tests/test_validation.py` — added 2 regression tests for `column_or_1d` and `check_array` with `StringDtype`
- `doc/whats_new/upcoming_changes/sklearn.utils/33383.fix.rst` — changelog entry
**Test results:**
- `sklearn/utils/tests/test_validation.py`: 306 passed, 115 skipped
- `sklearn/preprocessing/tests/test_label.py`: 55 passed, 99 skipped
#### AI usage disclosure
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
The changelog file is named `33383.fix.rst` — happy to rename it to the PR number once this is open if that's the convention (the README in that directory mentions using the PR number). | {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33451/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33451/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33451",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33451",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33451.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33451.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33450 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33450/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33450/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33450/events | https://github.com/scikit-learn/scikit-learn/pull/33450 | 4,018,794,981 | PR_kwDOAAzd1s7Ht3ID | 33,450 | DOC: clarify precisions_cholesky_ stores upper-triangular Cholesky factor | {
"login": "turanalmammadov",
"id": 16321061,
"node_id": "MDQ6VXNlcjE2MzIxMDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/16321061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/turanalmammadov",
"html_url": "https://github.com/turanalmammadov",
"followers_url": "https://api.github.com/users/turanalmammadov/followers",
"following_url": "https://api.github.com/users/turanalmammadov/following{/other_user}",
"gists_url": "https://api.github.com/users/turanalmammadov/gists{/gist_id}",
"starred_url": "https://api.github.com/users/turanalmammadov/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/turanalmammadov/subscriptions",
"organizations_url": "https://api.github.com/users/turanalmammadov/orgs",
"repos_url": "https://api.github.com/users/turanalmammadov/repos",
"events_url": "https://api.github.com/users/turanalmammadov/events{/privacy}",
"received_events_url": "https://api.github.com/users/turanalmammadov/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 1815445744,
"node_id": "MDU6TGFiZWwxODE1... | open | false | [] | null | 0 | 2026-03-03T21:30:07 | 2026-03-03T21:30:29 | null | null | NONE | null | null | null | null | ## What does this implement/fix? Explain your changes.
Closes #10141
The `precisions_cholesky_` attribute in `GaussianMixture` was described as:
> The Cholesky decomposition of the precision matrices of each mixture component.
This phrasing is ambiguous — it could mean the full factorization object, or the lower-triangular factor `L` such that `L @ L.T = Λ`.
In reality the code stores the **upper-triangular** factor `U = L^{-T}` (where `L = chol(Σ)`) such that `U @ U.T = Λ = Σ^{-1}`. This is documented precisely in the private helper `_compute_precision_cholesky_from_precisions` but was not reflected in the public attribute docstring or the `_compute_precision_cholesky` function docstring.
## Changes
- Updated `precisions_cholesky_` attribute docstring to explicitly state "upper-triangular Cholesky factor `U` such that `U @ U.T` equals the precision matrix `Λ`".
- Updated the `_compute_precision_cholesky` function docstring to describe the mathematical relationship `U = L^{-T}` (for full/tied covariances) and element-wise square root of precisions (for diag/spherical).
## Did you add a test? If not, why?
Documentation-only change. No new tests needed.
Made with [Cursor](https://cursor.com) | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33450/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33450/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33450",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33450",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33450.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33450.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33449 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33449/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33449/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33449/events | https://github.com/scikit-learn/scikit-learn/pull/33449 | 4,018,619,269 | PR_kwDOAAzd1s7HtSxr | 33,449 | FIX SGD partial_fit no-op on all-zero sample_weight | {
"login": "repeating",
"id": 37001921,
"node_id": "MDQ6VXNlcjM3MDAxOTIx",
"avatar_url": "https://avatars.githubusercontent.com/u/37001921?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/repeating",
"html_url": "https://github.com/repeating",
"followers_url": "https://api.github.com/users/repeating/followers",
"following_url": "https://api.github.com/users/repeating/following{/other_user}",
"gists_url": "https://api.github.com/users/repeating/gists{/gist_id}",
"starred_url": "https://api.github.com/users/repeating/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/repeating/subscriptions",
"organizations_url": "https://api.github.com/users/repeating/orgs",
"repos_url": "https://api.github.com/users/repeating/repos",
"events_url": "https://api.github.com/users/repeating/events{/privacy}",
"received_events_url": "https://api.github.com/users/repeating/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-03-03T20:43:38 | 2026-03-05T08:22:54 | 2026-03-05T08:22:54 | null | NONE | null | null | null | null | #### Reference Issues/PRs
Closes #33436.
#### What does this implement/fix? Explain your changes.
`SGDClassifier.partial_fit`, `SGDRegressor.partial_fit`, and `SGDOneClassSVM.partial_fit` currently raise `ValueError` when `sample_weight` is all zeros (behavior introduced in PR #32212). That's fine for `fit()` — fitting a fresh model on a zero-weight dataset is almost certainly a bug — but for `partial_fit` it's disruptive.
The use case is incremental/streaming workflows where zero-weight batches are used as a skip signal: you want to pass a batch through the pipeline unchanged without adding try/except blocks or pre-checks around every `partial_fit` call.
**The fix:** add an early-return in `_partial_fit` for all three estimators when all sample weights are zero. The guard fires *after* model state is initialized (so `coef_`, `intercept_`, `t_` are allocated on the first call) but *before* the SGD loop runs. This means `coef_`, `intercept_`, and `t_` are left completely unchanged.
`fit()` behavior is untouched — it still raises `ValueError` on all-zero weights.
Changes:
- `_stochastic_gradient.py`: added `_skip_on_zero_weights` parameter to `_partial_fit` in `BaseSGDClassifier`, `BaseSGDRegressor`, and `SGDOneClassSVM`. When called via `partial_fit`, this is `True`, which allows the zero-weight check to pass through `_check_sample_weight` and then triggers the early-return guard.
- `tests/test_sgd.py`: two new non-regression tests — one parametrized over `SGDClassifier`/`SGDRegressor`, one for `SGDOneClassSVM` — verifying that `coef_`, `intercept_`/`offset_`, and `t_` are unchanged after a zero-weight `partial_fit` call.
- `doc/whats_new/upcoming_changes/sklearn.linear_model/33436.fix.rst`: changelog entry.
#### AI usage disclosure
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
I verified that all 270 SGD tests pass and the 222 common estimator checks pass (including `check_all_zero_sample_weights_error`, which only tests `fit()` and is unaffected by this change). | {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33449/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33449/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33449",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33449",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33449.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33449.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33447 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33447/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33447/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33447/events | https://github.com/scikit-learn/scikit-learn/pull/33447 | 4,018,008,531 | PR_kwDOAAzd1s7HrS2F | 33,447 | FIX remove changelog entry merged in by mistake | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3730476391,
"node_id": "LA_kwDOAAzd1s7eWo1n",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Quick%20Review",
"name": "Quick Review",
"color": "79EEC4",
"default": false,
"description": "For PRs that are quick to review"
}
] | closed | false | [] | null | 0 | 2026-03-03T17:59:21 | 2026-03-03T23:55:33 | 2026-03-03T23:55:33 | null | MEMBER | null | null | null | null | I made the mistake of not re-checking the diff tab before clicking the merge button of #33437...
Sorry for that. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33447/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33447/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33447",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33447",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33447.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33447.patch",
"merged_at": "2026-03-03T23:55:33"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33446 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33446/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33446/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33446/events | https://github.com/scikit-learn/scikit-learn/issues/33446 | 4,017,963,813 | I_kwDOAAzd1s7vfUMl | 33,446 | ENH: Add FrequencyEncoder to sklearn.preprocessing | {
"login": "Sachin-MR05",
"id": 221297219,
"node_id": "U_kgDODTC6Qw",
"avatar_url": "https://avatars.githubusercontent.com/u/221297219?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Sachin-MR05",
"html_url": "https://github.com/Sachin-MR05",
"followers_url": "https://api.github.com/users/Sachin-MR05/followers",
"following_url": "https://api.github.com/users/Sachin-MR05/following{/other_user}",
"gists_url": "https://api.github.com/users/Sachin-MR05/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Sachin-MR05/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Sachin-MR05/subscriptions",
"organizations_url": "https://api.github.com/users/Sachin-MR05/orgs",
"repos_url": "https://api.github.com/users/Sachin-MR05/repos",
"events_url": "https://api.github.com/users/Sachin-MR05/events{/privacy}",
"received_events_url": "https://api.github.com/users/Sachin-MR05/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995756,
"node_id": "MDU6TGFiZWwxMjk5NTc1Ng==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/New%20Feature",
"name": "New Feature",
"color": "02e10c",
"default": false,
"description": null
},
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1... | closed | false | [] | null | 1 | 2026-03-03T17:49:50 | 2026-03-05T06:21:07 | 2026-03-05T06:21:07 | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the workflow you want to enable
When working with high-cardinality categorical columns (e.g. city names,
product IDs, user IDs), users need a simple, leak-proof way to encode
categories as their frequency in the training set.
Currently users must hand-roll this inside a pipeline or install external
libraries like category_encoders or feature-engine, which is error-prone
and adds unnecessary dependencies.
Example:
enc = FrequencyEncoder(normalize=True)
enc.fit_transform(X[['city']])
# London (50% of data) → 0.50
# Paris (33% of data) → 0.33
# Tokyo (17% of data) → 0.17
### Describe your proposed solution
Add a FrequencyEncoder transformer to sklearn.preprocessing with the
following API:
FrequencyEncoder(
normalize=True, # True = relative freq, False = raw count
handle_unknown=0.0, # value for unseen categories at transform time
min_frequency=None, # collapse rare labels below this threshold
)
Key design decisions:
- Works inside Pipeline and ColumnTransformer (no data leakage)
- Implements get_feature_names_out()
- Supports set_output(transform="pandas")
- Mirrors OrdinalEncoder conventions for consistency
- Multiple columns encoded independently in one transformer
### Describe alternatives you've considered, if relevant
_No response_
### Additional context
This encoding technique is widely used in tabular ML competitions (Kaggle)
and production pipelines. It is simple, target-free, and produces a single
compact column regardless of cardinality — making it more scalable than
OneHotEncoder for large categorical vocabularies.
I am willing to implement this if maintainers agree it belongs in core. | {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33446/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33446/timeline | null | not_planned | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33445 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33445/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33445/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33445/events | https://github.com/scikit-learn/scikit-learn/pull/33445 | 4,017,306,667 | PR_kwDOAAzd1s7Ho_hE | 33,445 | MAINT only run the cuda tests on the CUDA CI runner | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 4691539216,
"node_id": "LA_kwDOAAzd1s8AAAA... | closed | false | [] | null | 3 | 2026-03-03T15:29:54 | 2026-03-05T14:00:23 | 2026-03-05T13:22:41 | null | MEMBER | null | null | null | null | Since the number of array API related tests is growing quite quickly and since we already run all tests for non-CUDA devices on other CI hosts, I figured we could change the pytest filter for this config.
This will also allow use to have more freedom in the naming of test functions parametrized for array API namespace testing and as a result have more freedom to evolve our array API testing policy:
https://github.com/scikit-learn/scikit-learn/pull/33345/changes#r2845855747 | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33445/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33445/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33445",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33445",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33445.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33445.patch",
"merged_at": "2026-03-05T13:22:41"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33444 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33444/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33444/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33444/events | https://github.com/scikit-learn/scikit-learn/issues/33444 | 4,017,144,544 | I_kwDOAAzd1s7vcMLg | 33,444 | RFC a plan to move array API support out of experimental | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1314220606,
"node_id": "MDU6TGFiZWwxMzE0MjIwNjA2",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Needs%20Decision",
"name": "Needs Decision",
"color": "ffd8c1",
"default": false,
"description": "Requires decision"
},
{
"id": 1784583753,
"node_i... | open | false | [] | null | 10 | 2026-03-03T14:57:58 | 2026-03-06T11:02:06 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | Now that we have added array API support to several important models and tools (including classifiers, regressors, transformers, meta estimators, pipelines, evaluation and model selection tools...), it might be important to discuss what are our plans and conditions to remove the "experimental" marker for that set of features.
To me, moving out of experimental means the following things in particular:
- a. Setting `array_api_dispatch=True` as the default value in the config (or maybe even deprecate this configuration parameter to always have it enabled and simplify the code as a result);
- b. Enforcing backward compatibility guarantees with array API related behavior change (the requirement for 2 release deprecation cycle and the introduction of `FutureWarning`).
For a., I think we would need to:
- [ ] ensure that all estimators and public functions/classes that have not yet been updated to support array API inputs do not raise confusing error messages on non-(NumPy + CPU) inputs when array API dispatch is enabled. I have no idea if this is the case or not. We should probably extend the common test to check that.
Furthermore, for a. to make sense we might need to wait for SciPy to also enable array API support without having the user to manually set `SCIPY_ARRAY_API=1`, and therefore:
- [ ] coordinate with upstream on their own plans to move out of experimental.
I think that for point b. we at least need to implement the following:
- [x] finalize https://github.com/scikit-learn/scikit-learn/pull/32755 to ensure consistency in mixed namespace inputs for metric functions;
- [ ] at least the common testing infrastructure for https://github.com/scikit-learn/scikit-learn/issues/28668 (to make sure that we agree on the semantics) and possibly also upgrade all our existing estimators to ensure that they pass this new common test.
- [ ] https://github.com/scikit-learn/scikit-learn/pull/33076
- [ ] https://github.com/scikit-learn/scikit-learn/pull/32873
Once this is done, it means that we would also be able to consider an extra condition:
- [ ] at least one runnable example in our doc that shows how to use scikit-learn estimators in a non-trivial pipeline with on a non-CPU device (when available).
The purpose of this issue is to agree with a shared plan and maybe milestone this as a goal for 1.9?
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33444/reactions",
"total_count": 4,
"+1": 4,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33444/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33443 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33443/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33443/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33443/events | https://github.com/scikit-learn/scikit-learn/pull/33443 | 4,016,356,587 | PR_kwDOAAzd1s7Hl3K2 | 33,443 | DOC Improve description of values expected by `y_score` param | {
"login": "StefanieSenger",
"id": 91849487,
"node_id": "U_kgDOBXmDDw",
"avatar_url": "https://avatars.githubusercontent.com/u/91849487?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/StefanieSenger",
"html_url": "https://github.com/StefanieSenger",
"followers_url": "https://api.github.com/users/StefanieSenger/followers",
"following_url": "https://api.github.com/users/StefanieSenger/following{/other_user}",
"gists_url": "https://api.github.com/users/StefanieSenger/gists{/gist_id}",
"starred_url": "https://api.github.com/users/StefanieSenger/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/StefanieSenger/subscriptions",
"organizations_url": "https://api.github.com/users/StefanieSenger/orgs",
"repos_url": "https://api.github.com/users/StefanieSenger/repos",
"events_url": "https://api.github.com/users/StefanieSenger/events{/privacy}",
"received_events_url": "https://api.github.com/users/StefanieSenger/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 1815445423,
"node_id": "MDU6TGFiZWwxODE1... | open | false | [] | null | 2 | 2026-03-03T12:16:07 | 2026-03-05T12:51:51 | null | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
As discussed in https://github.com/scikit-learn/scikit-learn/pull/32732#discussion_r2866729326
#### What does this implement/fix? Explain your changes.
Improve description of values expected by `y_score` param, which listed three possibilities since #6336 of which two were meant the same.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [x] Research and understanding
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33443/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33443/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33443",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33443",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33443.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33443.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33442 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33442/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33442/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33442/events | https://github.com/scikit-learn/scikit-learn/pull/33442 | 4,015,398,238 | PR_kwDOAAzd1s7HiuXj | 33,442 | DOC Clarify HMM removal reasoning in FAQ | {
"login": "Aaditya-ops-cpu",
"id": 198593038,
"node_id": "U_kgDOC9ZKDg",
"avatar_url": "https://avatars.githubusercontent.com/u/198593038?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Aaditya-ops-cpu",
"html_url": "https://github.com/Aaditya-ops-cpu",
"followers_url": "https://api.github.com/users/Aaditya-ops-cpu/followers",
"following_url": "https://api.github.com/users/Aaditya-ops-cpu/following{/other_user}",
"gists_url": "https://api.github.com/users/Aaditya-ops-cpu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Aaditya-ops-cpu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Aaditya-ops-cpu/subscriptions",
"organizations_url": "https://api.github.com/users/Aaditya-ops-cpu/orgs",
"repos_url": "https://api.github.com/users/Aaditya-ops-cpu/repos",
"events_url": "https://api.github.com/users/Aaditya-ops-cpu/events{/privacy}",
"received_events_url": "https://api.github.com/users/Aaditya-ops-cpu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
}
] | closed | false | [] | null | 2 | 2026-03-03T08:33:54 | 2026-03-04T04:05:25 | 2026-03-04T00:16:29 | null | NONE | null | null | null | null | Clarified and guided to the reason for removing HMMs from scikit-learn and referenced design decision details.
<!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [✅] Documentation (including examples)
- [✅] Research and understanding
#### Any other comments?
The current one was a simple internal link. Adding a brief one-sentence explanation provides immediate context to the user before they click through to the more technical "Adding graphical models" section. This aligns with making the documentation more accessible to beginners.
Changes:
Modified doc/faq.rst to include a summary of why HMMs are out of scope.
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33442/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33442/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33442",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33442",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33442.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33442.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33441 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33441/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33441/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33441/events | https://github.com/scikit-learn/scikit-learn/pull/33441 | 4,014,639,030 | PR_kwDOAAzd1s7HgRsv | 33,441 | FIX Unsigned integer overflow in `RidgeClassifier` label binarisation | {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 1815448196,
"node_id": "MDU6... | open | false | [] | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/milestones/70",
"html_url": "https://github.com/scikit-learn/scikit-learn/milestone/70",
"labels_url": "https://api.github.com/repos/scikit-learn/scikit-learn/milestones/70/labels",
"id": 14414033,
"node_id": "MI_kwDOAAzd1s4A2_DR",
"number": 70,
"title": "1.8.1",
"description": "",
"creator": {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
},
"open_issues": 4,
"closed_issues": 5,
"state": "open",
"created_at": "2025-12-23T17:07:24",
"updated_at": "2026-03-03T04:28:26",
"due_on": "2026-03-03T00:00:00",
"closed_at": null
} | 0 | 2026-03-03T04:27:50 | 2026-03-03T06:01:33 | null | null | MEMBER | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Fixes #33390
#### What does this implement/fix? Explain your changes.
This PR fixes an overflow issue when [<code>RidgeClassifier</code>](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.RidgeClassifier.html) is fitted with unsigned integer labels.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [x] Research and understanding
#### Any other comments?
N / A
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33441/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33441/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33441",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33441",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33441.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33441.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33440 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33440/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33440/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33440/events | https://github.com/scikit-learn/scikit-learn/pull/33440 | 4,012,615,017 | PR_kwDOAAzd1s7HZsou | 33,440 | ENH add sparse and sample_weight support to MultiTaskElasticNet | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2911057044,
"node_id": "MDU6... | open | false | [] | null | 0 | 2026-03-02T18:12:04 | 2026-03-05T10:50:23 | null | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
Closes #3702, finally.
Contributes to #33457
#### What does this implement/fix? Explain your changes.
This add support to fit
- with `sample_weight`
- on sparse `X`
for
- `MultiTaskElasticNet`
- `MultiTaskLasso`
- `MultiTaskElasticNetCV`
- `MultiTaskLassoCV`
#### AI usage disclosure
None
#### Any other comments?
https://github.com/scikit-learn/scikit-learn/pull/15436#issuecomment-586288621:
> It's just that I hear something that we are consistent in linear models
with certain options implemented
only in certain estimators without clear reason. So it suggests we should
have sample_weights added to CV
classes very soon and ideally in the same release cycle
*Only* 6 years later and not in the same release 😄 (but note that the comment did not apply to multitask estimators of this PR).
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33440/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33440/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33440",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33440",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33440.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33440.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33439 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33439/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33439/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33439/events | https://github.com/scikit-learn/scikit-learn/pull/33439 | 4,012,344,801 | PR_kwDOAAzd1s7HYyiP | 33,439 | Fix standard scaler negative weights | {
"login": "DimiChatzipavlis",
"id": 165230559,
"node_id": "U_kgDOCdk33w",
"avatar_url": "https://avatars.githubusercontent.com/u/165230559?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DimiChatzipavlis",
"html_url": "https://github.com/DimiChatzipavlis",
"followers_url": "https://api.github.com/users/DimiChatzipavlis/followers",
"following_url": "https://api.github.com/users/DimiChatzipavlis/following{/other_user}",
"gists_url": "https://api.github.com/users/DimiChatzipavlis/gists{/gist_id}",
"starred_url": "https://api.github.com/users/DimiChatzipavlis/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/DimiChatzipavlis/subscriptions",
"organizations_url": "https://api.github.com/users/DimiChatzipavlis/orgs",
"repos_url": "https://api.github.com/users/DimiChatzipavlis/repos",
"events_url": "https://api.github.com/users/DimiChatzipavlis/events{/privacy}",
"received_events_url": "https://api.github.com/users/DimiChatzipavlis/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815448196,
"node_id": "MDU6TGFiZWwxODE1NDQ4MTk2",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:preprocessing",
"name": "module:preprocessing",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_id": "MD... | closed | false | [] | null | 1 | 2026-03-02T17:12:22 | 2026-03-03T04:16:52 | 2026-03-03T04:16:48 | null | NONE | null | null | null | null | #### What does this implement/fix?
`StandardScaler.fit` and `.partial_fit` now reject negative `sample_weight`.
They now call `_check_sample_weight(..., ensure_non_negative=True)`
This prevents negative weights from producing impossible negative variance, a RuntimeWarning, and silent `scale_=1` fallback.
#### Any other comments?
- Validation hardening: zero behavior change for valid (non-negative) weights.
- Added minimal test `test_standard_scaler_negative_sample_weight` covering both `fit` and `partial_fit`.
#### Notes
This is a gap in the preprocessing module (`sklearn.preprocessing`).
Most estimators are solid with the `ensure_non_negative` flag. Preprocessing has not this flag.
**AI usage disclosure**
I used AI assistance for:
- [x] Code generation
- [x] Test generation
| {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33439/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33439/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33439",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33439",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33439.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33439.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33438 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33438/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33438/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33438/events | https://github.com/scikit-learn/scikit-learn/issues/33438 | 4,011,805,560 | I_kwDOAAzd1s7vH0t4 | 33,438 | CI Add aggregating job in GHA to summarize statuses of all unit test jobs and be used a required check | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | open | false | [] | null | 0 | 2026-03-02T15:17:06 | 2026-03-02T15:22:10 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | Follow-up on https://github.com/scikit-learn/scikit-learn/issues/32434, the idea would be to have a single ggregating job (green if all unit test jobs are green, red if one of the unit test is red) to use it as required checks. Azure had a built-in aggregated job which we were using for our required check and right now our GHA required checks are based on a number of exact job names which is a bit brittle.
From a quick look some time ago https://github.com/scikit-learn/scikit-learn/pull/31832#issuecomment-3116678817:
- [github.com/orgs/community/discussions/4324#discussioncomment-13335158](https://github.com/orgs/community/discussions/4324#discussioncomment-13335158)
- third-party GitHub action [re-actors/alls-green](https://github.com/re-actors/alls-green?rgh-link-date=2025-07-25T07%3A08%3A55Z) that seems to handle this use case
I am wondering whether if this is doable/easy to take into account the status "optional" jobs, for example free-threaded or scipy-dev in a PR.
----------------
More details, GitHub required checks are managed centrally via the web UI on a job level and are based on exact job names. Setting is [here](https://github.com/scikit-learn/scikit-learn/settings/branch_protection_rules/23621449)
<img width="794" height="772" alt="Image" src="https://github.com/user-attachments/assets/2a29e0fd-2364-4eff-8037-f1374c395218" />
That means that it you change a job name in `main` and then update the required checks, all the PRs will have a pending job that never complete and the PR can not be merged. The work-around is to merge `main` into your PR branch. Having a single aggregating job used a
I think @FrancoisPgm had an initial look at one point, feel free to add more relevant context if you have some. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33438/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33438/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33437 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33437/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33437/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33437/events | https://github.com/scikit-learn/scikit-learn/pull/33437 | 4,010,550,006 | PR_kwDOAAzd1s7HS1W2 | 33,437 | FIX always pass `fill_value` to `xp.full` as per array API spec | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 5464227604,
"node_id": "LA_k... | closed | false | [] | null | 6 | 2026-03-02T10:42:05 | 2026-03-03T17:50:48 | 2026-03-03T17:48:50 | null | MEMBER | null | null | null | null | As discussed in https://github.com/data-apis/array-api-strict/issues/55, passing 0d arrays as `fill_value` to `xp.full` is not guaranteed to work for all array API libraries, hence `array-api-strict` started to reject such inputs as of version 2.5.
This PR should fix the failures discovered in the related lockfile update PRS: #33434 and #33432. | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33437/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33437/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33437",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33437",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33437.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33437.patch",
"merged_at": "2026-03-03T17:48:50"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33436 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33436/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33436/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33436/events | https://github.com/scikit-learn/scikit-learn/issues/33436 | 4,010,513,970 | I_kwDOAAzd1s7vC5Yy | 33,436 | SGDClassifier.partial_fit mutates the model when sample_weight is all zeros | {
"login": "yangbaechu",
"id": 45089264,
"node_id": "MDQ6VXNlcjQ1MDg5MjY0",
"avatar_url": "https://avatars.githubusercontent.com/u/45089264?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yangbaechu",
"html_url": "https://github.com/yangbaechu",
"followers_url": "https://api.github.com/users/yangbaechu/followers",
"following_url": "https://api.github.com/users/yangbaechu/following{/other_user}",
"gists_url": "https://api.github.com/users/yangbaechu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yangbaechu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yangbaechu/subscriptions",
"organizations_url": "https://api.github.com/users/yangbaechu/orgs",
"repos_url": "https://api.github.com/users/yangbaechu/repos",
"events_url": "https://api.github.com/users/yangbaechu/events{/privacy}",
"received_events_url": "https://api.github.com/users/yangbaechu/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995755,
"node_id": "MDU6TGFiZWwxMjk5NTc1NQ==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Bug",
"name": "Bug",
"color": "e10c02",
"default": false,
"description": null
},
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "ht... | open | false | [] | null | 1 | 2026-03-02T10:34:43 | 2026-03-03T20:43:19 | null | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the bug and give evidence about its user-facing impact
`SGDClassifier.partial_fit` changes the model state even when the new batch has `sample_weight=np.zeros(n_samples)`. In the code below, an all-zero-weight batch changes `coef_`, which affects subsequent predictions, and advances `t_`, which changes the learning-rate schedule for later partial_fit calls. I expect such a batch to behave like skipping it entirely. This matters to me because in weighted incremental-learning workflows, zero sample weights are a natural way to ignore samples or whole batches without changing the surrounding batching code. And the current behavior means that batches kept with zero weight are not equivalent to skipped batches, which silently changes the optimization path.
Related to #16298
### Steps/Code to Reproduce
```python
import numpy as np
from sklearn.datasets import make_classification
from sklearn.linear_model import SGDClassifier
def main():
X, y = make_classification(n_samples=40, random_state=1)
clf = SGDClassifier(random_state=0)
clf.partial_fit(X[:10], y[:10], classes=np.array([0, 1]))
coef_before = clf.coef_.copy()
t_before = clf.t_
clf.partial_fit(X[10:20], y[10:20], sample_weight=np.zeros(10))
coef_changed = bool(np.max(np.abs(clf.coef_ - coef_before)) > 0)
print("coef_changed:", coef_changed)
print("t_before:", t_before)
print("t_after:", clf.t_)
if __name__ == "__main__":
main()
```
### Expected Results
coef_changed: False
t_before: 11.0
t_after: 11.0
### Actual Results
coef_changed: True
t_before: 11.0
t_after: 21.0
### Versions
```shell
System:
python: 3.10.12 (main, Jan 26 2026, 14:55:28) [GCC 11.4.0]
executable: /usr/bin/python3.10
machine: Linux-6.6.87.2-microsoft-standard-WSL2-x86_64-with-glibc2.35
Python dependencies:
sklearn: 1.7.2
pip: 26.0.1
setuptools: 59.6.0
numpy: 2.2.6
scipy: 1.15.3
Cython: None
pandas: None
matplotlib: None
joblib: 1.5.3
threadpoolctl: 3.6.0
Built with OpenMP: True
threadpoolctl info:
user_api: blas
internal_api: openblas
num_threads: 8
prefix: libscipy_openblas
filepath: /home/yang/.local/lib/python3.10/site-packages/numpy.libs/libscipy_openblas64_-56d6093b.so
version: 0.3.29
threading_layer: pthreads
architecture: Haswell
user_api: blas
internal_api: openblas
num_threads: 8
prefix: libscipy_openblas
filepath: /home/yang/.local/lib/python3.10/site-packages/scipy.libs/libscipy_openblas-68440149.so
version: 0.3.28
threading_layer: pthreads
architecture: Haswell
user_api: openmp
internal_api: openmp
num_threads: 8
prefix: libgomp
filepath: /home/yang/.local/lib/python3.10/site-packages/scikit_learn.libs/libgomp-a34b3233.so.1.0.0
version: None
```
### Interest in fixing the bug
Yes, I would be interested in working on a PR if this is triaged as a bug. The issue seems to be that zero-weight samples are only zeroed out at the gradient-update stage, but are not skipped earlier in the optimization loop. The coefficient change disappears with `penalty=None`, which suggests the current behavior is tied to regularization being applied before zero-weight samples are fully skipped. My plan is to skip zero-weight samples earlier in the low-level SGD loop and add a test for an all-zero-weight `partial_fit` batch. Does that sound like the right direction? | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33436/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33436/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33435 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33435/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33435/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33435/events | https://github.com/scikit-learn/scikit-learn/pull/33435 | 4,009,415,328 | PR_kwDOAAzd1s7HPKYQ | 33,435 | CI Enable LLVM (clang-cl) wheel builds for Win ARM64 | {
"login": "dashsantosh-mcw",
"id": 158429614,
"node_id": "U_kgDOCXFxrg",
"avatar_url": "https://avatars.githubusercontent.com/u/158429614?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dashsantosh-mcw",
"html_url": "https://github.com/dashsantosh-mcw",
"followers_url": "https://api.github.com/users/dashsantosh-mcw/followers",
"following_url": "https://api.github.com/users/dashsantosh-mcw/following{/other_user}",
"gists_url": "https://api.github.com/users/dashsantosh-mcw/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dashsantosh-mcw/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dashsantosh-mcw/subscriptions",
"organizations_url": "https://api.github.com/users/dashsantosh-mcw/orgs",
"repos_url": "https://api.github.com/users/dashsantosh-mcw/repos",
"events_url": "https://api.github.com/users/dashsantosh-mcw/events{/privacy}",
"received_events_url": "https://api.github.com/users/dashsantosh-mcw/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 1815448196,
"node_id": "MDU6TGFiZWwxODE1ND... | closed | false | [] | null | 1 | 2026-03-02T05:46:04 | 2026-03-02T10:27:29 | 2026-03-02T10:27:28 | null | NONE | null | null | null | null | #### Reference Issues/PRs
N/A
#### What does this implement/fix? Explain your changes.
This PR enables `clang-cl` support for Windows ARM64 builds in the CI pipeline.
Majority of benchmarks show performance improvement with `clang-cl` (about ~50%).
Minor to low regressions observed in a few cases (upto ~3%).
Changes include:
- clang-cl based build support
- Fix for uint32_t redefinition on Windows with clang
- Updated libomp.dll handling in vendor.py
- Update `test_sizeof_LARGEST_INT_t` test to handle `clang-cl`
#### AI usage disclosure
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
### Benchmark Results
#### 1. cluster.KMeansBenchmark.time_predict
| Representation | Algorithm | MSVC (init_random) | Clang-CL (init_random) | Δ % | MSVC (init_kmeans++) | Clang-CL (init_kmeans++) | Δ % |
|----------------|-----------|-------------------|----------------------|-----|---------------------|------------------------|-----|
| dense | lloyd | 2.61 ms | 2.00 ms | +23.37% | 2.85 ms | 2.01 ms | +29.47% |
| dense | elkan | 2.84 ms | 2.03 ms | +28.52% | 2.76 ms | 2.04 ms | +26.09% |
| sparse | lloyd | 5.81 ms | 4.89 ms | +15.83% | 5.29 ms | 4.63 ms | +12.48% |
| sparse | elkan | 6.30 ms | 5.44 ms | +13.65% | 6.28 ms | 5.47 ms | +12.90% |
---
#### 2. cluster.KMeansBenchmark.time_transform
| Representation | Algorithm | MSVC (init_random) | Clang-CL (init_random) | Δ % | MSVC (init_kmeans++) | Clang-CL (init_kmeans++) | Δ % |
|----------------|-----------|-------------------|----------------------|-----|---------------------|------------------------|-----|
| dense | lloyd | 38.5 ms | 34.5 ms | +10.39% | 37.0 ms | 32.8 ms | +11.35% |
| dense | elkan | 33.6 ms | 32.7 ms | +2.68% | 31.6 ms | 32.1 ms | -1.58% |
| sparse | lloyd | 4890 ms | 5060 ms | -3.48% | 5030 ms | 5050 ms | -0.40% |
| sparse | elkan | 4950 ms | 5060 ms | -2.22% | 4980 ms | 5160 ms | -3.61% |
---
#### 3. linear_model.RidgeBenchmark.time_predict
| Representation | Solver | MSVC | Clang-CL | Δ % |
|----------------|--------|------|----------|-----|
| dense | auto | 14.6 ms | 10.3 ms | +29.45% |
| dense | svd | 10.4 ms | 10.0 ms | +3.85% |
| dense | cholesky | 9.96 ms | 10.1 ms | -1.41% |
| dense | lsqr | 18.4 ms | 10.0 ms | +45.65% |
| dense | sparse_cg | 10.2 ms | 10.2 ms | 0% |
| dense | sag | 14.8 ms | 10.2 ms | +31.08% |
| dense | saga | 10.1 ms | 10.3 ms | -1.98% |
| sparse | auto | 3.22 ms | 3.24 ms | -0.62% |
| sparse | cholesky | 3.27 ms | 3.25 ms | +0.61% |
| sparse | lsqr | 3.32 ms | 3.28 ms | +1.20% |
| sparse | sparse_cg | 3.22 ms | 3.24 ms | -0.62% |
| sparse | sag | 3.19 ms | 3.26 ms | -2.19% |
| sparse | saga | 3.21 ms | 3.22 ms | -0.31% |
---
#### 4. KNeighborsClassifierBenchmark.time_predict
| Algorithm | MSVC (low) | Clang-CL (low) | Δ % | MSVC (high) | Clang-CL (high) | Δ % |
|-----------|-----------|---------------|-----|------------|----------------|-----|
| brute | 89.0 ms | 89.5 ms | -0.56% | 4889 ms | 4678 ms | +4.32% |
| kd_tree | 308 ms | 308 ms | 0% | 5381 ms | 3152 ms | +41.42% |
| ball_tree | 761 ms | 535 ms | +29.70% | 5291 ms | 2622 ms | +50.44% |
| {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33435/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33435/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33435",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33435",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33435.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33435.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33434 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33434/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33434/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33434/events | https://github.com/scikit-learn/scikit-learn/pull/33434 | 4,009,363,266 | PR_kwDOAAzd1s7HO_su | 33,434 | :lock: :robot: CI Update lock files for main CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 2 | 2026-03-02T05:26:27 | 2026-03-02T15:58:21 | 2026-03-02T15:58:21 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33434/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33434/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33434",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33434",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33434.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33434.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33433 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33433/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33433/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33433/events | https://github.com/scikit-learn/scikit-learn/pull/33433 | 4,009,356,284 | PR_kwDOAAzd1s7HO-TV | 33,433 | :lock: :robot: CI Update lock files for free-threaded CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 0 | 2026-03-02T05:23:49 | 2026-03-02T10:20:21 | 2026-03-02T10:20:21 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33433/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33433/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33433",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33433",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33433.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33433.patch",
"merged_at": "2026-03-02T10:20:21"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33432 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33432/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33432/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33432/events | https://github.com/scikit-learn/scikit-learn/pull/33432 | 4,009,356,259 | PR_kwDOAAzd1s7HO-TB | 33,432 | :lock: :robot: CI Update lock files for array-api CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 5464227604,
"node_id": "LA_kwDOAAzd1s8AAAA... | open | false | [] | null | 2 | 2026-03-02T05:23:49 | 2026-03-02T10:18:44 | null | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33432/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33432/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33432",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33432",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33432.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33432.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33431 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33431/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33431/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33431/events | https://github.com/scikit-learn/scikit-learn/pull/33431 | 4,009,354,898 | PR_kwDOAAzd1s7HO-Bc | 33,431 | :lock: :robot: CI Update lock files for scipy-dev CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 0 | 2026-03-02T05:23:14 | 2026-03-02T10:19:56 | 2026-03-02T10:19:56 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33431/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33431/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33431",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33431",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33431.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33431.patch",
"merged_at": "2026-03-02T10:19:56"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33430 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33430/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33430/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33430/events | https://github.com/scikit-learn/scikit-learn/pull/33430 | 4,008,239,309 | PR_kwDOAAzd1s7HLfat | 33,430 | DOC: explain why __ is used instead of . for nested params (#33428) | {
"login": "Lavanyacheshani",
"id": 143724777,
"node_id": "U_kgDOCJEQ6Q",
"avatar_url": "https://avatars.githubusercontent.com/u/143724777?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/Lavanyacheshani",
"html_url": "https://github.com/Lavanyacheshani",
"followers_url": "https://api.github.com/users/Lavanyacheshani/followers",
"following_url": "https://api.github.com/users/Lavanyacheshani/following{/other_user}",
"gists_url": "https://api.github.com/users/Lavanyacheshani/gists{/gist_id}",
"starred_url": "https://api.github.com/users/Lavanyacheshani/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/Lavanyacheshani/subscriptions",
"organizations_url": "https://api.github.com/users/Lavanyacheshani/orgs",
"repos_url": "https://api.github.com/users/Lavanyacheshani/repos",
"events_url": "https://api.github.com/users/Lavanyacheshani/events{/privacy}",
"received_events_url": "https://api.github.com/users/Lavanyacheshani/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 1815446740,
"node_id": "MDU6TGFiZWwxODE1... | closed | false | [] | null | 0 | 2026-03-01T20:49:25 | 2026-03-03T04:44:14 | 2026-03-03T04:44:14 | null | NONE | null | null | null | null | ### What
Clarified why double underscore (`__`) is used instead of dot (`.`) for nested parameters in GridSearchCV.
Changes applied to:
- Glossary
- User Guide
- GridSearchCV docstring
### Why
The previous wording was unclear and could confuse users when setting parameters in a Pipeline.
This documentation update helps beginners understand the correct syntax.
### Notes
No code behavior is changed; this is purely documentation. | {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33430/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33430/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33430",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33430",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33430.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33430.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33429 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33429/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33429/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33429/events | https://github.com/scikit-learn/scikit-learn/pull/33429 | 4,007,338,319 | PR_kwDOAAzd1s7HIjWM | 33,429 | Bump the actions group with 2 updates | {
"login": "dependabot[bot]",
"id": 49699333,
"node_id": "MDM6Qm90NDk2OTkzMzM=",
"avatar_url": "https://avatars.githubusercontent.com/in/29110?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dependabot%5Bbot%5D",
"html_url": "https://github.com/apps/dependabot",
"followers_url": "https://api.github.com/users/dependabot%5Bbot%5D/followers",
"following_url": "https://api.github.com/users/dependabot%5Bbot%5D/following{/other_user}",
"gists_url": "https://api.github.com/users/dependabot%5Bbot%5D/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dependabot%5Bbot%5D/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dependabot%5Bbot%5D/subscriptions",
"organizations_url": "https://api.github.com/users/dependabot%5Bbot%5D/orgs",
"repos_url": "https://api.github.com/users/dependabot%5Bbot%5D/repos",
"events_url": "https://api.github.com/users/dependabot%5Bbot%5D/events{/privacy}",
"received_events_url": "https://api.github.com/users/dependabot%5Bbot%5D/received_events",
"type": "Bot",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5OD... | closed | false | [] | null | 3 | 2026-03-01T14:25:50 | 2026-03-03T11:36:25 | 2026-03-03T11:35:45 | null | CONTRIBUTOR | null | null | null | null | Bumps the actions group with 2 updates: [actions/download-artifact](https://github.com/actions/download-artifact) and [actions/upload-artifact](https://github.com/actions/upload-artifact).
Updates `actions/download-artifact` from 7 to 8
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/actions/download-artifact/releases">actions/download-artifact's releases</a>.</em></p>
<blockquote>
<h2>v8.0.0</h2>
<h2>v8 - What's new</h2>
<h3>Direct downloads</h3>
<p>To support direct uploads in <code>actions/upload-artifact</code>, the action will no longer attempt to unzip all downloaded files. Instead, the action checks the <code>Content-Type</code> header ahead of unzipping and skips non-zipped files. Callers wishing to download a zipped file as-is can also set the new <code>skip-decompress</code> parameter to <code>false</code>.</p>
<h3>Enforced checks (breaking)</h3>
<p>A previous release introduced digest checks on the download. If a download hash didn't match the expected hash from the server, the action would log a warning. Callers can now configure the behavior on mismatch with the <code>digest-mismatch</code> parameter. To be secure by default, we are now defaulting the behavior to <code>error</code> which will fail the workflow run.</p>
<h3>ESM</h3>
<p>To support new versions of the @actions/* packages, we've upgraded the package to ESM.</p>
<h2>What's Changed</h2>
<ul>
<li>Don't attempt to un-zip non-zipped downloads by <a href="https://github.com/danwkennedy"><code>@danwkennedy</code></a> in <a href="https://redirect.github.com/actions/download-artifact/pull/460">actions/download-artifact#460</a></li>
<li>Add a setting to specify what to do on hash mismatch and default it to <code>error</code> by <a href="https://github.com/danwkennedy"><code>@danwkennedy</code></a> in <a href="https://redirect.github.com/actions/download-artifact/pull/461">actions/download-artifact#461</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a href="https://github.com/actions/download-artifact/compare/v7...v8.0.0">https://github.com/actions/download-artifact/compare/v7...v8.0.0</a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="https://github.com/actions/download-artifact/commit/70fc10c6e5e1ce46ad2ea6f2b72d43f7d47b13c3"><code>70fc10c</code></a> Merge pull request <a href="https://redirect.github.com/actions/download-artifact/issues/461">#461</a> from actions/danwkennedy/digest-mismatch-behavior</li>
<li><a href="https://github.com/actions/download-artifact/commit/f258da9a506b755b84a09a531814700b86ccfc62"><code>f258da9</code></a> Add change docs</li>
<li><a href="https://github.com/actions/download-artifact/commit/ccc058e5fbb0bb2352213eaec3491e117cbc4a5c"><code>ccc058e</code></a> Fix linting issues</li>
<li><a href="https://github.com/actions/download-artifact/commit/bd7976ba57ecea96e6f3df575eb922d11a12a9fd"><code>bd7976b</code></a> Add a setting to specify what to do on hash mismatch and default it to <code>error</code></li>
<li><a href="https://github.com/actions/download-artifact/commit/ac21fcf45e0aaee541c0f7030558bdad38d77d6c"><code>ac21fcf</code></a> Merge pull request <a href="https://redirect.github.com/actions/download-artifact/issues/460">#460</a> from actions/danwkennedy/download-no-unzip</li>
<li><a href="https://github.com/actions/download-artifact/commit/15999bff51058bc7c19b50ebbba518eaef7c26c0"><code>15999bf</code></a> Add note about package bumps</li>
<li><a href="https://github.com/actions/download-artifact/commit/974686ed5098c7f9c9289ec946b9058e496a2561"><code>974686e</code></a> Bump the version to <code>v8</code> and add release notes</li>
<li><a href="https://github.com/actions/download-artifact/commit/fbe48b1d2756394be4cd4358ed3bc1343b330e75"><code>fbe48b1</code></a> Update test names to make it clearer what they do</li>
<li><a href="https://github.com/actions/download-artifact/commit/96bf374a614d4360e225874c3efd6893a3f285e7"><code>96bf374</code></a> One more test fix</li>
<li><a href="https://github.com/actions/download-artifact/commit/b8c4819ef592cbe04fd93534534b38f853864332"><code>b8c4819</code></a> Fix skip decompress test</li>
<li>Additional commits viewable in <a href="https://github.com/actions/download-artifact/compare/v7...v8">compare view</a></li>
</ul>
</details>
<br />
Updates `actions/upload-artifact` from 6 to 7
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/actions/upload-artifact/releases">actions/upload-artifact's releases</a>.</em></p>
<blockquote>
<h2>v7.0.0</h2>
<h2>v7 What's new</h2>
<h3>Direct Uploads</h3>
<p>Adds support for uploading single files directly (unzipped). Callers can set the new <code>archive</code> parameter to <code>false</code> to skip zipping the file during upload. Right now, we only support single files. The action will fail if the glob passed resolves to multiple files. The <code>name</code> parameter is also ignored with this setting. Instead, the name of the artifact will be the name of the uploaded file.</p>
<h3>ESM</h3>
<p>To support new versions of the <code>@actions/*</code> packages, we've upgraded the package to ESM.</p>
<h2>What's Changed</h2>
<ul>
<li>Add proxy integration test by <a href="https://github.com/Link"><code>@Link</code></a>- in <a href="https://redirect.github.com/actions/upload-artifact/pull/754">actions/upload-artifact#754</a></li>
<li>Upgrade the module to ESM and bump dependencies by <a href="https://github.com/danwkennedy"><code>@danwkennedy</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/762">actions/upload-artifact#762</a></li>
<li>Support direct file uploads by <a href="https://github.com/danwkennedy"><code>@danwkennedy</code></a> in <a href="https://redirect.github.com/actions/upload-artifact/pull/764">actions/upload-artifact#764</a></li>
</ul>
<h2>New Contributors</h2>
<ul>
<li><a href="https://github.com/Link"><code>@Link</code></a>- made their first contribution in <a href="https://redirect.github.com/actions/upload-artifact/pull/754">actions/upload-artifact#754</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a href="https://github.com/actions/upload-artifact/compare/v6...v7.0.0">https://github.com/actions/upload-artifact/compare/v6...v7.0.0</a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="https://github.com/actions/upload-artifact/commit/bbbca2ddaa5d8feaa63e36b76fdaad77386f024f"><code>bbbca2d</code></a> Support direct file uploads (<a href="https://redirect.github.com/actions/upload-artifact/issues/764">#764</a>)</li>
<li><a href="https://github.com/actions/upload-artifact/commit/589182c5a4cec8920b8c1bce3e2fab1c97a02296"><code>589182c</code></a> Upgrade the module to ESM and bump dependencies (<a href="https://redirect.github.com/actions/upload-artifact/issues/762">#762</a>)</li>
<li><a href="https://github.com/actions/upload-artifact/commit/47309c993abb98030a35d55ef7ff34b7fa1074b5"><code>47309c9</code></a> Merge pull request <a href="https://redirect.github.com/actions/upload-artifact/issues/754">#754</a> from actions/Link-/add-proxy-integration-tests</li>
<li><a href="https://github.com/actions/upload-artifact/commit/02a8460834e70dab0ce194c64360c59dc1475ef0"><code>02a8460</code></a> Add proxy integration test</li>
<li>See full diff in <a href="https://github.com/actions/upload-artifact/compare/v6...v7">compare view</a></li>
</ul>
</details>
<br />
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore <dependency name> major version` will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
- `@dependabot ignore <dependency name> minor version` will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
- `@dependabot ignore <dependency name>` will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
- `@dependabot unignore <dependency name>` will remove all of the ignore conditions of the specified dependency
- `@dependabot unignore <dependency name> <ignore condition>` will remove the ignore condition of the specified dependency and ignore conditions
</details> | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33429/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33429/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33429",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33429",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33429.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33429.patch",
"merged_at": "2026-03-03T11:35:45"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33428 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33428/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33428/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33428/events | https://github.com/scikit-learn/scikit-learn/issues/33428 | 4,007,036,814 | I_kwDOAAzd1s7u1oeO | 33,428 | Should we use dot (.) instead of double underscore (__) when using GridSearchCV with Pipeline | {
"login": "vietnh1009",
"id": 47221207,
"node_id": "MDQ6VXNlcjQ3MjIxMjA3",
"avatar_url": "https://avatars.githubusercontent.com/u/47221207?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/vietnh1009",
"html_url": "https://github.com/vietnh1009",
"followers_url": "https://api.github.com/users/vietnh1009/followers",
"following_url": "https://api.github.com/users/vietnh1009/following{/other_user}",
"gists_url": "https://api.github.com/users/vietnh1009/gists{/gist_id}",
"starred_url": "https://api.github.com/users/vietnh1009/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/vietnh1009/subscriptions",
"organizations_url": "https://api.github.com/users/vietnh1009/orgs",
"repos_url": "https://api.github.com/users/vietnh1009/repos",
"events_url": "https://api.github.com/users/vietnh1009/events{/privacy}",
"received_events_url": "https://api.github.com/users/vietnh1009/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Yb... | closed | false | [] | null | 0 | 2026-03-01T12:06:57 | 2026-03-02T00:35:54 | 2026-03-02T00:35:54 | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the issue linked to the documentation
Ok now I understand why
### Suggest a potential alternative/fix
No need | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33428/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33428/timeline | null | completed | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33427 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33427/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33427/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33427/events | https://github.com/scikit-learn/scikit-learn/pull/33427 | 4,006,666,861 | PR_kwDOAAzd1s7HGZ9n | 33,427 | FIX raise clear ValueError for mixed-type input in check_array | {
"login": "emmaeng700",
"id": 176187010,
"node_id": "U_kgDOCoBmgg",
"avatar_url": "https://avatars.githubusercontent.com/u/176187010?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/emmaeng700",
"html_url": "https://github.com/emmaeng700",
"followers_url": "https://api.github.com/users/emmaeng700/followers",
"following_url": "https://api.github.com/users/emmaeng700/following{/other_user}",
"gists_url": "https://api.github.com/users/emmaeng700/gists{/gist_id}",
"starred_url": "https://api.github.com/users/emmaeng700/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/emmaeng700/subscriptions",
"organizations_url": "https://api.github.com/users/emmaeng700/orgs",
"repos_url": "https://api.github.com/users/emmaeng700/repos",
"events_url": "https://api.github.com/users/emmaeng700/events{/privacy}",
"received_events_url": "https://api.github.com/users/emmaeng700/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815453311,
"node_id": "MDU6TGFiZWwxODE1NDUzMzEx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:utils",
"name": "module:utils",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-03-01T08:38:38 | 2026-03-02T00:35:29 | 2026-03-02T00:35:29 | null | NONE | null | null | null | null | When check_array receives mixed-type input that causes numpy to raise a cryptic ValueError, re-raise with a clear message explaining that mixed types are not supported. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33427/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33427/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33427",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33427",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33427.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33427.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33426 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33426/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33426/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33426/events | https://github.com/scikit-learn/scikit-learn/pull/33426 | 4,006,666,841 | PR_kwDOAAzd1s7HGZ9X | 33,426 | FIX avoid false DataConversionWarning in pairwise_distances_argmin with boolean metric | {
"login": "emmaeng700",
"id": 176187010,
"node_id": "U_kgDOCoBmgg",
"avatar_url": "https://avatars.githubusercontent.com/u/176187010?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/emmaeng700",
"html_url": "https://github.com/emmaeng700",
"followers_url": "https://api.github.com/users/emmaeng700/followers",
"following_url": "https://api.github.com/users/emmaeng700/following{/other_user}",
"gists_url": "https://api.github.com/users/emmaeng700/gists{/gist_id}",
"starred_url": "https://api.github.com/users/emmaeng700/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/emmaeng700/subscriptions",
"organizations_url": "https://api.github.com/users/emmaeng700/orgs",
"repos_url": "https://api.github.com/users/emmaeng700/repos",
"events_url": "https://api.github.com/users/emmaeng700/events{/privacy}",
"received_events_url": "https://api.github.com/users/emmaeng700/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815445423,
"node_id": "MDU6TGFiZWwxODE1NDQ1NDIz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:metrics",
"name": "module:metrics",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_id": "MDU6TGFiZWwyMz... | closed | false | [] | null | 1 | 2026-03-01T08:38:37 | 2026-03-02T00:34:06 | 2026-03-02T00:34:03 | null | NONE | null | null | null | null | Fixes #32495. When metric requires boolean data, pass dtype=bool to check_pairwise_arrays in pairwise_distances_argmin and pairwise_distances_argmin_min so the original bool dtype is preserved through the call chain and no spurious DataConversionWarning is emitted. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33426/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33426/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33426",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33426",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33426.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33426.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33425 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33425/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33425/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33425/events | https://github.com/scikit-learn/scikit-learn/pull/33425 | 4,005,756,628 | PR_kwDOAAzd1s7HDdQC | 33,425 | FIX preserve signed dtype in label_binarize for uint targets (#33390) | {
"login": "krishnakumarbhat",
"id": 79183768,
"node_id": "MDQ6VXNlcjc5MTgzNzY4",
"avatar_url": "https://avatars.githubusercontent.com/u/79183768?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/krishnakumarbhat",
"html_url": "https://github.com/krishnakumarbhat",
"followers_url": "https://api.github.com/users/krishnakumarbhat/followers",
"following_url": "https://api.github.com/users/krishnakumarbhat/following{/other_user}",
"gists_url": "https://api.github.com/users/krishnakumarbhat/gists{/gist_id}",
"starred_url": "https://api.github.com/users/krishnakumarbhat/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/krishnakumarbhat/subscriptions",
"organizations_url": "https://api.github.com/users/krishnakumarbhat/orgs",
"repos_url": "https://api.github.com/users/krishnakumarbhat/repos",
"events_url": "https://api.github.com/users/krishnakumarbhat/events{/privacy}",
"received_events_url": "https://api.github.com/users/krishnakumarbhat/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815448196,
"node_id": "MDU6TGFiZWwxODE1NDQ4MTk2",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:preprocessing",
"name": "module:preprocessing",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_id": "MD... | closed | false | [] | null | 0 | 2026-03-01T00:10:04 | 2026-03-01T08:04:05 | 2026-03-01T08:03:55 | null | NONE | null | null | null | null |
#### Reference Issues/PRs
Fixes #33390
#### What does this implement/fix? Explain your changes.
This PR fixes a numerical stability issue where `RidgeClassifier` produced incorrect (extreme) decision values when the input labels `y` were of an unsigned integer dtype (e.g., `uint64`).
**The Root Cause:**
In `sklearn/preprocessing/_label.py`, the `label_binarize` function was preserving the input's "integral" dtype. However, `RidgeClassifier` often uses a `neg_label` of `-1`. When `-1` is cast to an unsigned integer type (like `uint64`), it causes an integer overflow, resulting in a very large positive value (e.g., $2^{64}-1$).
**The Fix:**
I modified the dtype selection logic in `label_binarize` to only preserve the input dtype if it is a **signed integer**. If the input is unsigned, it now defaults to the system's standard indexing integer dtype (typically `int64`), which safely handles negative labels.
**Changes:**
* Updated `sklearn/preprocessing/_label.py` to use `xp.isdtype(y.dtype, "signed integer")`.
* Added a regression test in `sklearn/preprocessing/tests/test_label.py` that specifically checks for `uint64` inputs with negative labels.
#### AI usage disclosure
I used AI assistance for:
* [x] Code generation (e.g., when writing an implementation or fixing a bug)
* [ ] Test/benchmark generation
* [x] Documentation (including examples)
* [x] Research and understanding
#### Any other comments?
This fix ensures compatibility with the Python Array API standards by using explicit `isdtype` checks. I have verified that this restores parity between signed and unsigned integer inputs for `RidgeClassifier`.
| {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33425/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33425/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33425",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33425",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33425.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33425.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33424 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33424/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33424/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33424/events | https://github.com/scikit-learn/scikit-learn/pull/33424 | 4,004,865,653 | PR_kwDOAAzd1s7HBHy1 | 33,424 | ENH TimeSeriesSplit: add walk_forward rolling-origin mode | {
"login": "joelalcedo",
"id": 29354325,
"node_id": "MDQ6VXNlcjI5MzU0MzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/29354325?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/joelalcedo",
"html_url": "https://github.com/joelalcedo",
"followers_url": "https://api.github.com/users/joelalcedo/followers",
"following_url": "https://api.github.com/users/joelalcedo/following{/other_user}",
"gists_url": "https://api.github.com/users/joelalcedo/gists{/gist_id}",
"starred_url": "https://api.github.com/users/joelalcedo/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/joelalcedo/subscriptions",
"organizations_url": "https://api.github.com/users/joelalcedo/orgs",
"repos_url": "https://api.github.com/users/joelalcedo/repos",
"events_url": "https://api.github.com/users/joelalcedo/events{/privacy}",
"received_events_url": "https://api.github.com/users/joelalcedo/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815446740,
"node_id": "MDU6TGFiZWwxODE1NDQ2NzQw",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:model_selection",
"name": "module:model_selection",
"color": "151e91",
"default": false,
"description": ""
}
] | open | false | [] | null | 1 | 2026-02-28T15:43:01 | 2026-03-06T12:07:44 | null | null | NONE | null | null | null | null | ## Summary
This PR adds an **opt-in walk-forward / rolling-origin mode** to
`sklearn.model_selection.TimeSeriesSplit` via:
- `n_splits="walk_forward"` (sentinel mode)
- `min_train_size` to specify the initial training window size for expanding walk-forward evaluation
- `step` to control the stride between successive test starts
The goal is to support walk-forward evaluation workflows commonly used in time
series forecasting, while preserving existing `TimeSeriesSplit` behavior for
integer `n_splits`.
## Motivation
`TimeSeriesSplit` is currently primarily **`n_splits`-driven** ("choose number of
folds, then infer fold boundaries"). A common alternative workflow is
**window-size-driven** ("choose initial train length, test horizon, stride, and
derive how many splits exist from the data length").
This PR provides that window-size-driven workflow as an explicit opt-in mode via
`n_splits="walk_forward"`.
## API
### Expanding (cumulative) walk-forward
```python
from sklearn.model_selection import TimeSeriesSplit
cv = TimeSeriesSplit(
n_splits="walk_forward",
min_train_size=1260,
test_size=1,
step=21,
gap=0,
)
```
### Rolling (fixed-size) train window walk-forward
```python
from sklearn.model_selection import TimeSeriesSplit
cv = TimeSeriesSplit(
n_splits="walk_forward",
max_train_size=1260,
test_size=1,
step=21,
gap=0,
)
```
## Walk-forward mode semantics
In `n_splits="walk_forward"` mode:
- `test_size` defines the width of each test window.
- `gap` keeps its existing meaning: number of samples excluded between the end
of train and start of test.
- `step` defines the stride between successive test starts (defaults to 1 if not
provided).
- The training window is:
- **expanding** when `max_train_size=None` (uses `min_train_size` to define the
first train window), or
- **rolling** when `max_train_size` is provided (fixed-size training window).
Note: depending on `step` and `test_size`, successive test windows may overlap.
## Validation / parameter rules (walk_forward)
- `test_size` must be specified.
- `step` must be a positive integer (defaults to 1 if not provided).
- `gap` must be a non-negative integer.
- Exactly one of `min_train_size` or `max_train_size` must be provided:
- expanding mode: `min_train_size` is required and `max_train_size=None`
- rolling mode: `max_train_size` is required and `min_train_size=None`
In `n_splits="walk_forward"` mode, the number of splits depends on `n_samples`,
so `get_n_splits(X)` uses `X` to infer the number of feasible splits.
## Backward compatibility
- Existing behavior for integer `n_splits` remains unchanged.
- Walk-forward semantics are opt-in via `n_splits="walk_forward"`.
- `step` and `min_train_size` are only valid in walk-forward mode.
This uses a sentinel `n_splits="walk_forward"` opt-in to avoid ambiguity and
preserve the existing integer `n_splits` semantics, consistent with prior
discussion in #23780.
## Example usage
Expanding walk-forward example:
```python
import numpy as np
from sklearn.model_selection import TimeSeriesSplit
X = np.arange(20)
splitter = TimeSeriesSplit(
n_splits="walk_forward",
min_train_size=5,
test_size=2,
gap=1,
step=3,
)
print("Number of splits:", splitter.get_n_splits(X))
for i, (train_idx, test_idx) in enumerate(splitter.split(X)):
print(
f"Split {i}: "
f"train [{train_idx[0]}:{train_idx[-1]+1}], "
f"test [{test_idx[0]}:{test_idx[-1]+1}]"
)
if i >= 3:
print("...")
break
```
Expected output:
```text
Number of splits: 5
Split 0: train [0:5], test [6:8]
Split 1: train [0:8], test [9:11]
Split 2: train [0:11], test [12:14]
Split 3: train [0:14], test [15:17]
...
```
## Related issues / prior discussion
- Addresses #22523 (rolling/sliding / walk-forward time series splitting):
https://github.com/scikit-learn/scikit-learn/issues/22523
- Addresses #23923 (need to specify initial train size / rolling-origin semantics):
https://github.com/scikit-learn/scikit-learn/issues/23923
- Addresses #24243 (skip/stride between splits):
https://github.com/scikit-learn/scikit-learn/issues/24243
- Related prior PR/discussion:
https://github.com/scikit-learn/scikit-learn/pull/23780
## Tests / checks run locally
- Ruff (feature python files):
```bash
mamba run -n sklearn-dev python -m ruff check \
sklearn/model_selection/_split.py \
sklearn/model_selection/tests/test_split.py
```
- Targeted model_selection tests:
```bash
mamba run -n sklearn-dev python -m pytest \
sklearn/model_selection/tests/test_split.py \
-k "TimeSeriesSplit or walk_forward" -vv
```
- Pre-commit on branch diff vs upstream:
```bash
mamba run -n sklearn-dev pre-commit run --from-ref upstream/main --to-ref HEAD
```
## Documentation
- Updated `TimeSeriesSplit` documentation and examples to include walk-forward usage.
- Added user guide documentation for walk-forward evaluation.
- Added a whats_new entry for the new functionality. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33424/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33424/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33424",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33424",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33424.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33424.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33423 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33423/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33423/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33423/events | https://github.com/scikit-learn/scikit-learn/pull/33423 | 4,002,115,143 | PR_kwDOAAzd1s7G4aGD | 33,423 | [Callbacks] DOC Add callbacks developer documentation | {
"login": "FrancoisPgm",
"id": 35327799,
"node_id": "MDQ6VXNlcjM1MzI3Nzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/35327799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FrancoisPgm",
"html_url": "https://github.com/FrancoisPgm",
"followers_url": "https://api.github.com/users/FrancoisPgm/followers",
"following_url": "https://api.github.com/users/FrancoisPgm/following{/other_user}",
"gists_url": "https://api.github.com/users/FrancoisPgm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FrancoisPgm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FrancoisPgm/subscriptions",
"organizations_url": "https://api.github.com/users/FrancoisPgm/orgs",
"repos_url": "https://api.github.com/users/FrancoisPgm/repos",
"events_url": "https://api.github.com/users/FrancoisPgm/events{/privacy}",
"received_events_url": "https://api.github.com/users/FrancoisPgm/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5... | open | false | [] | null | 0 | 2026-02-27T17:36:44 | 2026-03-04T18:52:46 | null | null | CONTRIBUTOR | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Towards #27676
#### What does this implement/fix? Explain your changes.
Add a callback page in the developer documentation introducing callbacks, explaining the task tree and the callback context object.
Add an example showing how to make a custom estimator or meta-estimator compatible with callbacks.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
This PR corresponds to the PRs https://github.com/jeremiedbb/scikit-learn/pull/31 and https://github.com/jeremiedbb/scikit-learn/pull/27 that were targeting the fork's branch which was merged into the callbacks branch.
ping @jeremiedbb @StefanieSenger
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33423/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33423/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33423",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33423",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33423.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33423.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33422 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33422/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33422/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33422/events | https://github.com/scikit-learn/scikit-learn/pull/33422 | 4,001,837,961 | PR_kwDOAAzd1s7G3fk3 | 33,422 | FIX RidgeClassifier support for unsigned integer targets | {
"login": "shivansh023023",
"id": 168457476,
"node_id": "U_kgDOCgp1BA",
"avatar_url": "https://avatars.githubusercontent.com/u/168457476?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shivansh023023",
"html_url": "https://github.com/shivansh023023",
"followers_url": "https://api.github.com/users/shivansh023023/followers",
"following_url": "https://api.github.com/users/shivansh023023/following{/other_user}",
"gists_url": "https://api.github.com/users/shivansh023023/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shivansh023023/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shivansh023023/subscriptions",
"organizations_url": "https://api.github.com/users/shivansh023023/orgs",
"repos_url": "https://api.github.com/users/shivansh023023/repos",
"events_url": "https://api.github.com/users/shivansh023023/events{/privacy}",
"received_events_url": "https://api.github.com/users/shivansh023023/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_id": "MDU6... | closed | false | [] | null | 0 | 2026-02-27T16:28:34 | 2026-02-28T17:07:44 | 2026-02-28T17:07:31 | null | NONE | null | null | null | null | Reference Issues/PRs
Fixes #33390.
What does this implement/fix? Explain your changes.
This PR ensures that RidgeClassifier can handle target arrays (y) with unsigned integer dtypes (e.g., uint8, uint64).
Although some environments might handle these types implicitly, scikit-learn's internal LabelBinarizer and underlying C-extensions can be sensitive to unsigned types. I have added an explicit cast to int64 within the fit method of RidgeClassifier when an unsigned dtype is detected.
Changes included:
sklearn/linear_model/_ridge.py: Added a check for unsigned integer dtypes in RidgeClassifier.fit to cast y to np.int64.
sklearn/linear_model/tests/test_ridge.py: Added a new regression test test_ridge_classifier_unsigned_target to ensure various unsigned types are supported and produce correct prediction dtypes.
AI usage disclosure
I used AI assistance for:
[x] Code generation (e.g., when writing an implementation or fixing a bug)
[x] Test/benchmark generation
[ ] Documentation (including examples)
[x] Research and understanding
Any other comments?
The new test was verified using pytest and the file was formatted using ruff. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33422/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33422/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33422",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33422",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33422.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33422.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33421 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33421/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33421/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33421/events | https://github.com/scikit-learn/scikit-learn/pull/33421 | 4,001,671,740 | PR_kwDOAAzd1s7G28Oc | 33,421 | [Callbacks] Alternative to #33340 to cloning with callbacks | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 10245205856,
"node_id": ... | open | false | [] | null | 0 | 2026-02-27T15:48:42 | 2026-03-04T18:43:52 | null | null | MEMBER | null | null | null | null | As discussed in the last meeting this PR explores an alternative to https://github.com/scikit-learn/scikit-learn/pull/33340
ping @ogrisel @StefanieSenger @lesteve
Personally I tend to prefer slightly the other version because it doesn't extend the API of `clone`, but the coupling in the other PR is not ideal either.
(side note: the added test doesn't work yet because it needs https://github.com/scikit-learn/scikit-learn/pull/33394) | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33421/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33421/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33421",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33421",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33421.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33421.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33420 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33420/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33420/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33420/events | https://github.com/scikit-learn/scikit-learn/pull/33420 | 4,001,292,791 | PR_kwDOAAzd1s7G1sc0 | 33,420 | [Callbacks] Rename propagate_callbacks into propagate_callback_context | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 3730476391,
"node_id": "... | closed | false | [] | null | 0 | 2026-02-27T14:27:29 | 2026-03-02T10:56:54 | 2026-03-02T10:56:54 | null | MEMBER | null | null | null | null | to be more explicit about what this method does because it propagates the context and the callbacks.
Extracted from https://github.com/scikit-learn/scikit-learn/pull/33340 because we decided to not couple propagate_callback_context and clone.
ping @lesteve @StefanieSenger | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33420/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33420/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33420",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33420",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33420.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33420.patch",
"merged_at": "2026-03-02T10:56:54"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33419 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33419/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33419/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33419/events | https://github.com/scikit-learn/scikit-learn/pull/33419 | 4,001,043,988 | PR_kwDOAAzd1s7G03h2 | 33,419 | MNT fix and refactor color handling in DecisionBoundaryDisplay - Part 1 | {
"login": "AnneBeyer",
"id": 12881170,
"node_id": "MDQ6VXNlcjEyODgxMTcw",
"avatar_url": "https://avatars.githubusercontent.com/u/12881170?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AnneBeyer",
"html_url": "https://github.com/AnneBeyer",
"followers_url": "https://api.github.com/users/AnneBeyer/followers",
"following_url": "https://api.github.com/users/AnneBeyer/following{/other_user}",
"gists_url": "https://api.github.com/users/AnneBeyer/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AnneBeyer/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AnneBeyer/subscriptions",
"organizations_url": "https://api.github.com/users/AnneBeyer/orgs",
"repos_url": "https://api.github.com/users/AnneBeyer/repos",
"events_url": "https://api.github.com/users/AnneBeyer/events{/privacy}",
"received_events_url": "https://api.github.com/users/AnneBeyer/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444240,
"node_id": "MDU6TGFiZWwxODE1NDQ0MjQw",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:inspection",
"name": "module:inspection",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 1815453311,
"node_id": "MDU6TGFi... | closed | false | [] | null | 1 | 2026-02-27T13:28:28 | 2026-03-04T10:26:03 | 2026-03-03T13:03:55 | null | CONTRIBUTOR | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Towards #33115
Specifically, this implements checkboxes 1, 2, part of 3, and 5 (see next section for details)
#### What does this implement/fix? Explain your changes.
This PR moves all parts related to color selection based on 'multiclass_colors' into a sepparate function (3,5), unifies and simplifies the color selection mechanism (2) and adds a check that enough distinct colors are available in the given colormap (1).
It also adds a non-regression test for the last point, which is the only functionality change.
See also the comments in the code for more details.
@ogrisel @lucyleeow @StefanieSenger @antoinebaker
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [x] Research and understanding
#### Any other comments?
Unrelated small change:
I converted the error messages in `sklearn/utils/_optional_dependencies.py` to f-strings.
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33419/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33419/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33419",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33419",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33419.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33419.patch",
"merged_at": "2026-03-03T13:03:54"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33418 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33418/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33418/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33418/events | https://github.com/scikit-learn/scikit-learn/issues/33418 | 3,999,983,866 | I_kwDOAAzd1s7uauj6 | 33,418 | Deploy ASV benchmarks again | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 705180250,
"node_id": "MDU6TGFiZWw3MDUxODA... | open | false | [] | null | 1 | 2026-02-27T09:08:44 | 2026-03-02T10:07:13 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | In #16723 and #17026 we deployed airspeed velocity (asv) benchmarks. For a while now, they are offline. This is unfortunate because:
- It is harder to detect regressions.
- We don’t see the big time improvements in the last releases.
This issue is to track the (re-) deployment of the ASV benchmarks until they are back online.
@smarie @jeremiedbb @thomasjpfan @ogrisel @jnothman ping as you have been involved in the original setup (ossue or pr). | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33418/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33418/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33417 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33417/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33417/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33417/events | https://github.com/scikit-learn/scikit-learn/pull/33417 | 3,999,577,943 | PR_kwDOAAzd1s7GwASr | 33,417 | DOC clarify Nystroem precomputed-kernel input shapes | {
"login": "LarytheLord",
"id": 169234180,
"node_id": "U_kgDOChZPBA",
"avatar_url": "https://avatars.githubusercontent.com/u/169234180?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LarytheLord",
"html_url": "https://github.com/LarytheLord",
"followers_url": "https://api.github.com/users/LarytheLord/followers",
"following_url": "https://api.github.com/users/LarytheLord/following{/other_user}",
"gists_url": "https://api.github.com/users/LarytheLord/gists{/gist_id}",
"starred_url": "https://api.github.com/users/LarytheLord/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/LarytheLord/subscriptions",
"organizations_url": "https://api.github.com/users/LarytheLord/orgs",
"repos_url": "https://api.github.com/users/LarytheLord/repos",
"events_url": "https://api.github.com/users/LarytheLord/events{/privacy}",
"received_events_url": "https://api.github.com/users/LarytheLord/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
}
] | open | false | [] | null | 0 | 2026-02-27T07:20:02 | 2026-02-27T07:20:24 | null | null | NONE | null | null | null | null | #### Reference Issues/PRs
Closes #19178.
#### What does this implement/fix? Explain your changes.
This updates `Nystroem` documentation to remove ambiguity around `kernel="precomputed"`:
- documents that `X` is interpreted as a kernel matrix when `kernel="precomputed"`
- documents the current `fit` constraint (`n_components == n_samples`)
- clarifies `fit`/`transform` input shapes for precomputed-kernel usage
This is a docs-only clarification; no runtime behavior changes.
#### Any other comments?
Validation run locally:
- `python3 -m compileall sklearn/kernel_approximation.py`
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33417/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33417/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33417",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33417",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33417.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33417.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33416 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33416/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33416/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33416/events | https://github.com/scikit-learn/scikit-learn/issues/33416 | 3,998,497,462 | I_kwDOAAzd1s7uVDq2 | 33,416 | New algorithm for multi-label confusion matrix | {
"login": "almeidava93",
"id": 88211872,
"node_id": "MDQ6VXNlcjg4MjExODcy",
"avatar_url": "https://avatars.githubusercontent.com/u/88211872?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/almeidava93",
"html_url": "https://github.com/almeidava93",
"followers_url": "https://api.github.com/users/almeidava93/followers",
"following_url": "https://api.github.com/users/almeidava93/following{/other_user}",
"gists_url": "https://api.github.com/users/almeidava93/gists{/gist_id}",
"starred_url": "https://api.github.com/users/almeidava93/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/almeidava93/subscriptions",
"organizations_url": "https://api.github.com/users/almeidava93/orgs",
"repos_url": "https://api.github.com/users/almeidava93/repos",
"events_url": "https://api.github.com/users/almeidava93/events{/privacy}",
"received_events_url": "https://api.github.com/users/almeidava93/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995756,
"node_id": "MDU6TGFiZWwxMjk5NTc1Ng==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/New%20Feature",
"name": "New Feature",
"color": "02e10c",
"default": false,
"description": null
},
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1... | open | false | [] | null | 0 | 2026-02-27T01:02:04 | 2026-02-27T01:10:02 | null | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the workflow you want to enable
I would like to suggest the inclusion of a new confusion matrix algorithm that complies with scikit-learn’s inclusion criteria and addresses an important limitation of the one-vs-rest 2x2 confusion matrices currently implemented for multi-label classification problems in scikit-learn.
The paper that introduces this algorithm is the following:
```
@ARTICLE{9711932,
author={Heydarian, Mohammadreza and Doyle, Thomas E. and Samavi, Reza},
journal={IEEE Access},
title={MLCM: Multi-Label Confusion Matrix},
year={2022},
volume={10},
number={},
pages={19083-19095},
keywords={Measurement;Classification algorithms;Motion pictures;Loss measurement;Libraries;Task analysis;TV;Classification performance;confusion matrix;machine learning;multi-class;multi-label},
doi={10.1109/ACCESS.2022.3151048}
}
```
The paper introduces the Multi-Label Confusion Matrix (MLCM), a novel method for constructing a true 2-dimensional confusion matrix for multi-label classification tasks, where each instance may have multiple true and predicted labels. Unlike existing one-vs-rest approaches, which treat each label independently and can produce ambiguous or inconsistent counts of false positives and false negatives, MLCM systematically accounts for all combinations of true and predicted labels, including cases with no predicted labels (NPL) or no true labels (NTL). This framework preserves the interpretability and structural properties of the traditional multi-class confusion matrix, enables correct and balanced accounting of TP, FP, FN, and TN across classes, and allows precise computation of derived metrics such as precision, recall, and F-score without the distortions introduced by aggregate one-vs-rest methods.
I would like to suggest adding an option to the `sklearn.metrics.multilabel_confusion_matrix` function to allow selection of the computation method. I propose introducing a parameter called `mode`, with `pairwise` as the default (corresponding to the currently implemented algorithm), and an additional `aggregate` mode that implements the technique proposed in the paper.
The current pairwise confusion matrix has a limitation: it does not allow understanding which labels are being confused with one another. In contrast, the proposed method enables inspection of every label and how they are confused with other labels. This is particularly valuable for reporting classifier results and for analyzing classifier limitations.
**scikit-learn’s inclusion criteria**:
- Paper published in 2022
- More than 200 citations
- Clear utility
### Describe your proposed solution
I'm available to submiting the PR for this issue if the maintainers agree.
### Describe alternatives you've considered, if relevant
_No response_
### Additional context
_No response_ | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33416/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33416/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33415 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33415/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33415/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33415/events | https://github.com/scikit-learn/scikit-learn/pull/33415 | 3,996,525,260 | PR_kwDOAAzd1s7GmBfk | 33,415 | DOC Clarify max_features heuristics for Forest classes | {
"login": "shivansh023023",
"id": 168457476,
"node_id": "U_kgDOCgp1BA",
"avatar_url": "https://avatars.githubusercontent.com/u/168457476?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shivansh023023",
"html_url": "https://github.com/shivansh023023",
"followers_url": "https://api.github.com/users/shivansh023023/followers",
"following_url": "https://api.github.com/users/shivansh023023/following{/other_user}",
"gists_url": "https://api.github.com/users/shivansh023023/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shivansh023023/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shivansh023023/subscriptions",
"organizations_url": "https://api.github.com/users/shivansh023023/orgs",
"repos_url": "https://api.github.com/users/shivansh023023/repos",
"events_url": "https://api.github.com/users/shivansh023023/events{/privacy}",
"received_events_url": "https://api.github.com/users/shivansh023023/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 1815440243,
"node_id": "MDU6TGFiZWwxODE1... | open | false | [] | null | 1 | 2026-02-26T16:31:13 | 2026-02-26T16:55:54 | null | null | NONE | null | null | null | null | Reference Issues/PRs
Fixes #33386.
What does this implement/fix? Explain your changes.
This PR improves the documentation for RandomForestClassifier, RandomForestRegressor, and ExtraTreesClassifier by clarifying the heuristics behind the default max_features values.
Specifically, it adds explanations to the docstrings to help users understand:
Classification: Why "sqrt" is the default (reducing correlation between trees to improve ensemble performance).
Regression: Why 1.0 (all features) is the default (minimizing the bias of individual trees), while noting that smaller values can be used to reduce variance.
These changes address user confusion regarding the differing default behaviors between classification and regression tasks.
AI usage disclosure
I used AI assistance for:
[ ] Code generation (e.g., when writing an implementation or fixing a bug)
[ ] Test/benchmark generation
[x] Documentation (including examples)
[x] Research and understanding
Any other comments?
This is a documentation-only fix. I have verified the formatting using ruff. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33415/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33415/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33415",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33415",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33415.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33415.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33414 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33414/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33414/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33414/events | https://github.com/scikit-learn/scikit-learn/pull/33414 | 3,995,887,998 | PR_kwDOAAzd1s7Gj6W8 | 33,414 | TST improve test time in `test_calibrated_classifier_cv_works_with_large_confidence_scores` | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 3730476391,
"node_id": "... | closed | false | [] | null | 0 | 2026-02-26T14:26:37 | 2026-03-02T16:31:59 | 2026-03-02T16:31:54 | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
None
#### What does this implement/fix? Explain your changes.
In some CI runs `test_calibrated_classifier_cv_works_with_large_confidence_scores` is the slowest test.
#### AI usage disclosure
None | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33414/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33414/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33414",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33414",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33414.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33414.patch",
"merged_at": "2026-03-02T16:31:54"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33413 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33413/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33413/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33413/events | https://github.com/scikit-learn/scikit-learn/pull/33413 | 3,995,148,509 | PR_kwDOAAzd1s7GhcXy | 33,413 | FIX avoid repeated warning emission in Parallel (no-hyperopt reproducible) | {
"login": "gufarui",
"id": 205157738,
"node_id": "U_kgDODDp1ag",
"avatar_url": "https://avatars.githubusercontent.com/u/205157738?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gufarui",
"html_url": "https://github.com/gufarui",
"followers_url": "https://api.github.com/users/gufarui/followers",
"following_url": "https://api.github.com/users/gufarui/following{/other_user}",
"gists_url": "https://api.github.com/users/gufarui/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gufarui/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gufarui/subscriptions",
"organizations_url": "https://api.github.com/users/gufarui/orgs",
"repos_url": "https://api.github.com/users/gufarui/repos",
"events_url": "https://api.github.com/users/gufarui/events{/privacy}",
"received_events_url": "https://api.github.com/users/gufarui/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815453311,
"node_id": "MDU6TGFiZWwxODE1NDUzMzEx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:utils",
"name": "module:utils",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_id": "MDU6TGFiZWwyMzk5Mz... | closed | false | [] | null | 0 | 2026-02-26T12:05:46 | 2026-03-04T00:26:33 | 2026-03-04T00:26:29 | null | NONE | null | null | null | null | ## Summary
This PR keeps the same fix as #33257, but documents and validates a reproducer **without hyperopt**. @lesteve
Related to #33230.
Supersedes #33257.
The core change is in `sklearn.utils.parallel` warning-filter propagation:
- serialize warning categories as `(module, qualname)` references,
- restore from already-loaded modules only,
- avoid implicit third-party imports during worker-side unpickling.
## No-hyperopt reproducer
Reproducer uses only:
- `sklearn.utils.parallel.Parallel`
- Python stdlib
- `setuptools/pkg_resources` (no hyperopt)
```python
import time
import pkg_resources
from sklearn.utils.parallel import Parallel, delayed
Parallel(n_jobs=2, backend='loky')(delayed(time.sleep)(0) for _ in range(6))
print('DONE')
```
## Real outputs
### scikit-learn 1.8.0 (before fix)
```text
/tmp/repro_pkg_resources_only.py:2: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
import pkg_resources
/opt/homebrew/Cellar/python@3.12/3.12.12/Frameworks/Python.framework/Versions/3.12/lib/python3.12/multiprocessing/queues.py:122: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
return _ForkingPickler.loads(res)
/opt/homebrew/Cellar/python@3.12/3.12.12/Frameworks/Python.framework/Versions/3.12/lib/python3.12/multiprocessing/queues.py:122: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
return _ForkingPickler.loads(res)
DONE
```
Counts:
- `pkg_resources is deprecated as an API`: **3**
- `multiprocessing/queues.py:122`: **2**
### this PR branch (after fix)
```text
/tmp/repro_pkg_resources_only.py:2: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
import pkg_resources
DONE
```
Counts:
- `pkg_resources is deprecated as an API`: **1**
- `multiprocessing/queues.py:122`: **0**
### control: joblib.Parallel on scikit-learn 1.8.0
```text
/tmp/repro_pkg_resources_joblib.py:2: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
import pkg_resources
DONE
```
Count:
- `pkg_resources is deprecated as an API`: **1**
## Validation
Locally run:
```bash
python -m ruff check sklearn/utils/parallel.py sklearn/utils/tests/test_parallel.py
python -m pytest --color=no sklearn/utils/tests/test_parallel.py -q
python -m pytest --color=no sklearn/inspection/tests/test_permutation_importance.py -q
```
All passed locally.
---
## Attention
AI was used to correct the english grammer and sentence
| {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33413/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33413/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33413",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33413",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33413.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33413.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33412 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33412/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33412/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33412/events | https://github.com/scikit-learn/scikit-learn/issues/33412 | 3,994,754,948 | I_kwDOAAzd1s7uGx-E | 33,412 | RFC add Gradient Boosting with any estimator, deprecate AdaBoost | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995756,
"node_id": "MDU6TGFiZWwxMjk5NTc1Ng==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/New%20Feature",
"name": "New Feature",
"color": "02e10c",
"default": false,
"description": null
},
{
"id": 1314220606,
"node_id": "MDU6TGFiZWwxMzE0Mj... | open | false | [] | null | 2 | 2026-02-26T10:46:15 | 2026-03-03T08:10:54 | null | null | MEMBER | {
"id": 108050,
"node_id": "IT_kwDOAAWUPs4AAaYS",
"name": "Feature",
"description": "A request, idea, or new functionality",
"color": "blue",
"created_at": "2024-01-25T09:12:21",
"updated_at": "2024-10-08T08:30:56",
"is_enabled": true
} | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Gradient Boosting
As of scikit-learn version 1.8, we have
- `GradientBoostingClassifier`
- `GradientBoostingRegressor`
that perform gradient boosting with trees (`DecisionTreeRegressor`) as base estimator. What is missing is the ability to perform gradient boosting with **any base estimator**.
#### Proposal 1:
Add possibility for gradient boosting with arbitrary (regression) base learners.
### AdaBoost
AdaBoost (Freud & Shapire 1996) was once very successful and we have
- `AdaBoostClassifier`
- `AdaBoostRegressor`
Since its invention it has been shown, see [FHT2000](https://doi.org/10.1214/aos/1016218223) and [Friedman2001](https://doi.org/10.1214/aos/1013203451):
- AdaBoost is equivalent to Newton boosting with the exponential loss
- Discrete AdaBoost / AdaBoost.M1 (base learner predicts -1, +1, `predict`), the version we implemented for classifiers, seems to be inferior to Real AdaBoost (baser learner predicts probabilities, `predict_proba`), see FHT2000 Section 6.
#### Proposal 2:
If proposal 1 is implemented, then deprecate and remove AdaBoost estimators (and private `BaseWeightBoosting`).
This has a light dependence on #27873, which also suggests Newton boosting (for trees). | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33412/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33412/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33411 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33411/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33411/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33411/events | https://github.com/scikit-learn/scikit-learn/pull/33411 | 3,994,624,339 | PR_kwDOAAzd1s7GfroA | 33,411 | [Callbacks] FIX Always store the parent context in `propagate_callbacks` | {
"login": "FrancoisPgm",
"id": 35327799,
"node_id": "MDQ6VXNlcjM1MzI3Nzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/35327799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FrancoisPgm",
"html_url": "https://github.com/FrancoisPgm",
"followers_url": "https://api.github.com/users/FrancoisPgm/followers",
"following_url": "https://api.github.com/users/FrancoisPgm/following{/other_user}",
"gists_url": "https://api.github.com/users/FrancoisPgm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FrancoisPgm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FrancoisPgm/subscriptions",
"organizations_url": "https://api.github.com/users/FrancoisPgm/orgs",
"repos_url": "https://api.github.com/users/FrancoisPgm/repos",
"events_url": "https://api.github.com/users/FrancoisPgm/events{/privacy}",
"received_events_url": "https://api.github.com/users/FrancoisPgm/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-02-26T10:19:38 | 2026-03-02T11:01:38 | 2026-02-26T11:19:02 | null | CONTRIBUTOR | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Towards #27676
#### What does this implement/fix? Explain your changes.
In `propagate_callbacks`, the parent context is now always stored as an attribute of the sub estimator, even if there is no callback to propagate, as the sub-estimator might hold a non auto-propagated callback that will need to reconstruct the task tree.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
ping @jeremiedbb
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33411/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33411/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33411",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33411",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33411.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33411.patch",
"merged_at": "2026-02-26T11:19:02"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33410 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33410/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33410/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33410/events | https://github.com/scikit-learn/scikit-learn/pull/33410 | 3,992,827,439 | PR_kwDOAAzd1s7GZzQI | 33,410 | fix: replace 2 bare except clauses with except Exception | {
"login": "haosenwang1018",
"id": 167664334,
"node_id": "U_kgDOCf5azg",
"avatar_url": "https://avatars.githubusercontent.com/u/167664334?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/haosenwang1018",
"html_url": "https://github.com/haosenwang1018",
"followers_url": "https://api.github.com/users/haosenwang1018/followers",
"following_url": "https://api.github.com/users/haosenwang1018/following{/other_user}",
"gists_url": "https://api.github.com/users/haosenwang1018/gists{/gist_id}",
"starred_url": "https://api.github.com/users/haosenwang1018/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/haosenwang1018/subscriptions",
"organizations_url": "https://api.github.com/users/haosenwang1018/orgs",
"repos_url": "https://api.github.com/users/haosenwang1018/repos",
"events_url": "https://api.github.com/users/haosenwang1018/events{/privacy}",
"received_events_url": "https://api.github.com/users/haosenwang1018/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | closed | false | [] | null | 1 | 2026-02-26T02:05:33 | 2026-03-02T16:37:25 | 2026-03-02T16:37:25 | null | NONE | null | null | null | null | Replace bare except clauses with except Exception. | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33410/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33410/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33410",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33410",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33410.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33410.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33409 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33409/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33409/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33409/events | https://github.com/scikit-learn/scikit-learn/pull/33409 | 3,991,555,148 | PR_kwDOAAzd1s7GVlKX | 33,409 | ENH Add argument to SelectFwe to choose between Bonferroni and Holm method | {
"login": "mmbannert",
"id": 7759365,
"node_id": "MDQ6VXNlcjc3NTkzNjU=",
"avatar_url": "https://avatars.githubusercontent.com/u/7759365?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mmbannert",
"html_url": "https://github.com/mmbannert",
"followers_url": "https://api.github.com/users/mmbannert/followers",
"following_url": "https://api.github.com/users/mmbannert/following{/other_user}",
"gists_url": "https://api.github.com/users/mmbannert/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mmbannert/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mmbannert/subscriptions",
"organizations_url": "https://api.github.com/users/mmbannert/orgs",
"repos_url": "https://api.github.com/users/mmbannert/repos",
"events_url": "https://api.github.com/users/mmbannert/events{/privacy}",
"received_events_url": "https://api.github.com/users/mmbannert/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815442108,
"node_id": "MDU6TGFiZWwxODE1NDQyMTA4",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:feature_selection",
"name": "module:feature_selection",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-02-25T20:14:46 | 2026-03-05T09:08:49 | 2026-03-05T09:08:49 | null | NONE | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
##### Background
Feature selection is frequently used to reduce the dimensionality of large datasets. To this end, a univariate scoring function can be used to test if a feature correlates significantly with a variable of interest. However, testing a large group of features introduces the multiple comparisons problem where the type I error probability (i.e., the probability of erroneously rejecting a null hypothesis) exceeds the nominal error probability for the whole test family due to the large number of tests performed. A feature may thus appear to correlate with the target variable merely due to noise and end up getting selected erroneously.
`SelectFwe` implements the Bonferroni correction which ensures that the actual type I error probability corresponds to the nominal error probability by dividing the significance threshold for the _p_ value by the number of tests. This procedure, however, comes with the risk of increasing the type II error probability, i.e., the error of failing to reject the null hypothesis when it is in fact false. This may result in a failure to detect features although they are informative in reality.
The [Holm method](https://en.wikipedia.org/wiki/Holm–Bonferroni_method) (a.k.a. Holm-Bonferroni method) solves this problem by providing the same protection against type I errors as Bonferroni while reducing the probability of a type II error. In other words, it controls the family-wise error rate as effectively as the Bonferroni correction but is more sensitive to informative features. Although the Holm method is more powerful than Bonferroni, it unfortunately is not currently implemented in `SelectFwe`.
##### Solution
I modified `SelectFwe` so that it now accepts an additional input argument called `fwe_control` of type `str`. The user can choose between `bonf` for Bonferroni, which is its default setting for better backward compatibility, and `holm` for the Holm method. The Holm method sorts all _m_ null hypotheses of a test family by their corresponding _p_ values in ascending order $p_1 < p_2 < ... < p_m$ and applies a modified Bonferroni correction in a step-wise fashion. It rejects the first null hypothesis if $p_1 < \frac{\alpha}{m}$. If this null hypothesis is rejected, it rejects the second null hypothesis if $p_2 < \frac{\alpha}{m - 1}$. Generally, all subsequent null hypotheses are rejected if $p_i < \frac{\alpha}{m - i + 1}$ until a null hypothesis cannot be rejected anymore. All remaining null hypotheses are retained.
##### Demonstration
I verified that the Holm method is more powerful using the example that is provided in the documentation for `SelectFwe`.
```
import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import load_breast_cancer
from sklearn.feature_selection import SelectFwe, chi2
ALPHAS = np.array([.001, .002, .005, .01, .02, .05, .1, .2])
X, y = load_breast_cancer(return_X_y=True)
X.shape
print(f"Original number of features: {X.shape[1]}")
n_sig_feat = np.zeros_like(ALPHAS, dtype=int)
n_sig_feat_holm = np.zeros_like(ALPHAS, dtype=int)
for thresh_num, alpha in enumerate(ALPHAS):
X_new = SelectFwe(chi2, alpha=alpha).fit_transform(X, y)
n_sig_feat[thresh_num] = X_new.shape[1]
X_new_holm = SelectFwe(chi2, alpha=alpha, fwe_control="holm").fit_transform(X, y)
n_sig_feat_holm[thresh_num] = X_new_holm.shape[1]
plt.plot(ALPHAS, n_sig_feat, marker="o", label="Bonferroni")
plt.plot(ALPHAS, n_sig_feat_holm, marker="s", label="Holm")
plt.xscale("log")
plt.xlabel("alpha")
plt.ylabel("number of significant features")
plt.legend()
plt.show()
```
<img width="585" height="437" alt="image" src="https://github.com/user-attachments/assets/3168f91e-9d7f-4a2c-b3ef-2945dde38113" />
The example shows that the Holm method controls the family-wise error rate while detecting more informative features compared to the Bonferroni correction. The difference between the two measures is expected to scale with the number of input features.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [x] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
I tried to implement the new procedure with minimal changes to the existing code. Since the method has been well established in statistics and unchanged for a long time, I expect it to require only little maintenance in the future as well. I verified that the tests completed without errors after applying these changes. Using the demonstration shown above, I provide an example of how the new functionality is used and show that it behaves as expected. I am aware that non-regression tests should be carried out for new features. Unfortunately, I do not know what exactly a non-regression would entail in this case and how (or where) I should add it to the existing scikit-learn code. Help is very welcome.
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "virchan",
"id": 25701849,
"node_id": "MDQ6VXNlcjI1NzAxODQ5",
"avatar_url": "https://avatars.githubusercontent.com/u/25701849?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/virchan",
"html_url": "https://github.com/virchan",
"followers_url": "https://api.github.com/users/virchan/followers",
"following_url": "https://api.github.com/users/virchan/following{/other_user}",
"gists_url": "https://api.github.com/users/virchan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/virchan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/virchan/subscriptions",
"organizations_url": "https://api.github.com/users/virchan/orgs",
"repos_url": "https://api.github.com/users/virchan/repos",
"events_url": "https://api.github.com/users/virchan/events{/privacy}",
"received_events_url": "https://api.github.com/users/virchan/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33409/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33409/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33409",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33409",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33409.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33409.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33408 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33408/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33408/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33408/events | https://github.com/scikit-learn/scikit-learn/pull/33408 | 3,990,634,490 | PR_kwDOAAzd1s7GSe9X | 33,408 | [Callbacks] EFF Single manager for all callbacks | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 3730476391,
"node_id": "... | closed | false | [] | null | 0 | 2026-02-25T17:07:12 | 2026-02-27T13:12:16 | 2026-02-27T13:12:16 | null | MEMBER | null | null | null | null | Closes https://github.com/scikit-learn/scikit-learn/issues/33325
Create a sklearn global dedicated manager for all callbacks.
ping @FrancoisPgm @StefanieSenger @lesteve @ogrisel | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33408/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33408/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33408",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33408",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33408.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33408.patch",
"merged_at": "2026-02-27T13:12:16"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33407 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33407/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33407/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33407/events | https://github.com/scikit-learn/scikit-learn/pull/33407 | 3,990,211,358 | PR_kwDOAAzd1s7GRIJp | 33,407 | [Callbacks] FEA Add the MetricMonitor callback | {
"login": "FrancoisPgm",
"id": 35327799,
"node_id": "MDQ6VXNlcjM1MzI3Nzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/35327799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FrancoisPgm",
"html_url": "https://github.com/FrancoisPgm",
"followers_url": "https://api.github.com/users/FrancoisPgm/followers",
"following_url": "https://api.github.com/users/FrancoisPgm/following{/other_user}",
"gists_url": "https://api.github.com/users/FrancoisPgm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FrancoisPgm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FrancoisPgm/subscriptions",
"organizations_url": "https://api.github.com/users/FrancoisPgm/orgs",
"repos_url": "https://api.github.com/users/FrancoisPgm/repos",
"events_url": "https://api.github.com/users/FrancoisPgm/events{/privacy}",
"received_events_url": "https://api.github.com/users/FrancoisPgm/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 10245205856,
"node_id": ... | open | false | [] | null | 2 | 2026-02-25T15:41:44 | 2026-03-05T13:58:27 | null | null | CONTRIBUTOR | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Towards #27676
#### What does this implement/fix? Explain your changes.
Add the `MetricMonitor` callback, which evaluates and log a metric on an estimator's prediction during fit.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [x] Research and understanding
#### Any other comments?
ping @jeremiedbb
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33407/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33407/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33407",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33407",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33407.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33407.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33405 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33405/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33405/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33405/events | https://github.com/scikit-learn/scikit-learn/pull/33405 | 3,988,854,015 | PR_kwDOAAzd1s7GMqya | 33,405 | Fix wrong chance_level plot in PrecisionRecallDisplay when using Pytorch data | {
"login": "lucolivi",
"id": 10436644,
"node_id": "MDQ6VXNlcjEwNDM2NjQ0",
"avatar_url": "https://avatars.githubusercontent.com/u/10436644?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucolivi",
"html_url": "https://github.com/lucolivi",
"followers_url": "https://api.github.com/users/lucolivi/followers",
"following_url": "https://api.github.com/users/lucolivi/following{/other_user}",
"gists_url": "https://api.github.com/users/lucolivi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucolivi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucolivi/subscriptions",
"organizations_url": "https://api.github.com/users/lucolivi/orgs",
"repos_url": "https://api.github.com/users/lucolivi/repos",
"events_url": "https://api.github.com/users/lucolivi/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucolivi/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815445423,
"node_id": "MDU6TGFiZWwxODE1NDQ1NDIz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:metrics",
"name": "module:metrics",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 4691539216,
"node_id": "LA_kwDOAAzd1s8... | open | false | [] | null | 1 | 2026-02-25T11:20:54 | 2026-03-05T14:03:07 | null | null | NONE | null | null | null | null |
<!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Fixes #33342
#### What does this implement/fix? Explain your changes.
This fix the bug that caused the wrong chance level line to be plot in `PrecisionRecallDisplay.from_predictions` when `y_true` was a pytorch array. It was caused by the use of a `Counter` object to calculate the chance level, that makes use of hash functions to index values, which are not fixed for a given value in pytorch data types. Now the implementation compares `y_true` entries to `pos_label` values directly, without requiring `Counter` objects.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
No
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33405/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33405/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33405",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33405",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33405.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33405.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33404 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33404/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33404/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33404/events | https://github.com/scikit-learn/scikit-learn/pull/33404 | 3,988,837,948 | PR_kwDOAAzd1s7GMnXi | 33,404 | [Callbacks] POC Callback API for functions | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 10245205856,
"node_id": "LA_kwDOAAzd1s8AAAACYqlvYA",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Callbacks",
"name": "Callbacks",
"color": "7800F5",
"default": false,
"description": ""
}
] | open | false | [] | null | 2 | 2026-02-25T11:17:53 | 2026-02-26T16:01:57 | null | null | MEMBER | null | null | null | null | Related to https://github.com/scikit-learn/scikit-learn/pull/28760#discussion_r2818073843
This PR is POC for an extension of the callback API to work in functions that act like meta-estimators such as `cross_validate`.
This PR implements a minimal extension that does not handle nesting such functions, but I think that it should already cover most use cases.
This proposed extension add new callback hooks to be used in functions. Its main advantage is that it's completely separated from the API for estimators (see the diff: only additions, no modifications) while fully compatible: we use the same task tree and callbacks can be propagated from the function to estimators.
Given the clean separation, I think that it allows us to not rush that in the ongoing SLEP and to implement that in a future release as an api extension.
try this working example at home:
```py
from sklearn.callback.tests._utils import MaxIterEstimator
from sklearn.callback.tests.test_callback_support import my_function
from sklearn.callback import ProgressBar
my_function(MaxIterEstimator(), n_fits=10, n_jobs=2, callbacks=ProgressBar())
```
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33404/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33404/timeline | null | null | null | true | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33404",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33404",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33404.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33404.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33403 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33403/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33403/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33403/events | https://github.com/scikit-learn/scikit-learn/pull/33403 | 3,988,343,653 | PR_kwDOAAzd1s7GK9fw | 33,403 | CI Clean-up after Azure -> GHA move | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5OD... | closed | false | [] | null | 1 | 2026-02-25T09:43:32 | 2026-03-02T13:43:55 | 2026-03-02T13:43:28 | null | MEMBER | null | null | null | null | Clean-up mentioned in https://github.com/scikit-learn/scikit-learn/issues/32434#issuecomment-3921197837
I double-checked that there was no Azure related thing via the `git grep` below. The remaining files that mentions Azure:
- pip-compile `_lock.txt` lock-files that mentions the previous `build_tools/azure` in a comment. I will let the automatic update-lock file handle this next Monday.
- some Azure related packages in conda lock-files like azure-storage-common, I think this happens in environments that have pyarrow
```
❯ git grep -i azure
build_tools/github/debian_32bit_lock.txt:# pip-compile --output-file=build_tools/azure/debian_32bit_lock.txt build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/debian_32bit_lock.txt: # via -r build_tools/azure/debian_32bit_requirements.txt
build_tools/github/pylatest_conda_forge_cuda_array-api_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-core-cpp-1.16.2-h206d751_0.conda#5492abf806c45298ae642831c670bba0
build_tools/github/pylatest_conda_forge_cuda_array-api_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-identity-cpp-1.13.3-hed0cdb0_1.conda#68bfb556bdf56d56e9f38da696e752ca
build_tools/github/pylatest_conda_forge_cuda_array-api_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-storage-common-cpp-12.12.0-ha7a2c86_1.conda#6400f73fe5ebe19fe7aca3616f1f1de7
build_tools/github/pylatest_conda_forge_cuda_array-api_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-storage-blobs-cpp-12.16.0-hdd73cc9_1.conda#939d9ce324e51961c7c4c0046733dbb7
build_tools/github/pylatest_conda_forge_cuda_array-api_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-storage-files-datalake-cpp-12.14.0-h52c5a47_1.conda#6d10339800840562b7dad7775f5d2c16
build_tools/github/pylatest_conda_forge_mkl_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-core-cpp-1.16.2-h206d751_0.conda#5492abf806c45298ae642831c670bba0
build_tools/github/pylatest_conda_forge_mkl_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-identity-cpp-1.13.3-hed0cdb0_1.conda#68bfb556bdf56d56e9f38da696e752ca
build_tools/github/pylatest_conda_forge_mkl_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-storage-common-cpp-12.12.0-ha7a2c86_1.conda#6400f73fe5ebe19fe7aca3616f1f1de7
build_tools/github/pylatest_conda_forge_mkl_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-storage-blobs-cpp-12.16.0-hdd73cc9_1.conda#939d9ce324e51961c7c4c0046733dbb7
build_tools/github/pylatest_conda_forge_mkl_linux-64_conda.lock:https://conda.anaconda.org/conda-forge/linux-64/azure-storage-files-datalake-cpp-12.14.0-h52c5a47_1.conda#6d10339800840562b7dad7775f5d2c16
build_tools/github/ubuntu_atlas_lock.txt:# pip-compile --output-file=build_tools/azure/ubuntu_atlas_lock.txt build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # via -r build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # via -r build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # via -r build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # via -r build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # -r build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # via -r build_tools/azure/ubuntu_atlas_requirements.txt
build_tools/github/ubuntu_atlas_lock.txt: # via -r build_tools/azure/ubuntu_atlas_requirements.txt
``` | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33403/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33403/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33403",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33403",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33403.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33403.patch",
"merged_at": "2026-03-02T13:43:28"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33402 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33402/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33402/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33402/events | https://github.com/scikit-learn/scikit-learn/issues/33402 | 3,986,980,426 | I_kwDOAAzd1s7tpH5K | 33,402 | ⚠️ CI failed on Unit tests Linux x86-64 pymin_conda_forge_openblas_ubuntu_2204 (last failure: Feb 25, 2026) ⚠️ | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [] | open | false | [] | null | 3 | 2026-02-25T03:28:38 | 2026-03-06T03:29:07 | null | null | CONTRIBUTOR | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | **CI failed on [Unit tests Linux x86-64 pymin_conda_forge_openblas_ubuntu_2204](https://github.com/scikit-learn/scikit-learn/actions/runs/22380837654/job/64781087867)** (Feb 25, 2026)
Unable to find junit file. Please see link for details. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33402/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33402/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33401 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33401/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33401/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33401/events | https://github.com/scikit-learn/scikit-learn/issues/33401 | 3,985,786,083 | I_kwDOAAzd1s7tkkTj | 33,401 | Add "auto" option as default solver to GLMs | {
"login": "lorentzenchr",
"id": 15324633,
"node_id": "MDQ6VXNlcjE1MzI0NjMz",
"avatar_url": "https://avatars.githubusercontent.com/u/15324633?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lorentzenchr",
"html_url": "https://github.com/lorentzenchr",
"followers_url": "https://api.github.com/users/lorentzenchr/followers",
"following_url": "https://api.github.com/users/lorentzenchr/following{/other_user}",
"gists_url": "https://api.github.com/users/lorentzenchr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lorentzenchr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lorentzenchr/subscriptions",
"organizations_url": "https://api.github.com/users/lorentzenchr/orgs",
"repos_url": "https://api.github.com/users/lorentzenchr/repos",
"events_url": "https://api.github.com/users/lorentzenchr/events{/privacy}",
"received_events_url": "https://api.github.com/users/lorentzenchr/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Needs%20Triage",
"name": "Needs Triage",
"color": "FF0000",
"default": false,
"description": "Issue requires triage"
}
] | open | false | [] | null | 0 | 2026-02-24T20:57:20 | 2026-02-27T15:50:03 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | Currently, our GLMS like `LogisticRegression`, `PoissonRegressor` and others have default `solver="lbfgs"`. Unfortunately, this solver cannot solve L1-penalized objectives. Therefore, `LogisticRegression(l1_ratio=1)` errors.
I propose to change the default solver to `"auto"` with the meaning:
1. The solver chosen may change from version to version. (But this should happen rarely.)
2. The solver chosen is fit for the purpose no matter which panalty is chosen.
3. Start with
- No penalty and L2 penalties: `"lbfgs"`
- L1 and enet: `"saga"`, or, if #33160 gets merged first, one of the new `"newton-cd"` methods | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33401/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33401/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33400 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33400/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33400/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33400/events | https://github.com/scikit-learn/scikit-learn/pull/33400 | 3,984,939,827 | PR_kwDOAAzd1s7F_ynw | 33,400 | DOC clarify max_features defaults for RandomForestClassifier and RandomForrestRegressor. | {
"login": "srikrishp",
"id": 167325499,
"node_id": "U_kgDOCfkvOw",
"avatar_url": "https://avatars.githubusercontent.com/u/167325499?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/srikrishp",
"html_url": "https://github.com/srikrishp",
"followers_url": "https://api.github.com/users/srikrishp/followers",
"following_url": "https://api.github.com/users/srikrishp/following{/other_user}",
"gists_url": "https://api.github.com/users/srikrishp/gists{/gist_id}",
"starred_url": "https://api.github.com/users/srikrishp/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/srikrishp/subscriptions",
"organizations_url": "https://api.github.com/users/srikrishp/orgs",
"repos_url": "https://api.github.com/users/srikrishp/repos",
"events_url": "https://api.github.com/users/srikrishp/events{/privacy}",
"received_events_url": "https://api.github.com/users/srikrishp/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
}
] | closed | false | [] | null | 1 | 2026-02-24T17:30:06 | 2026-03-04T14:28:05 | 2026-03-04T14:28:04 | null | NONE | null | null | null | null | …omForestRegressor
<!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Closes #3308
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
This PR clarifies why the default value of 'max_features' differs between
RandomForestClassifier and RandomForestRegressor and briefly explains
the bias-variance tradeoff.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "ArturoAmorQ",
"id": 86408019,
"node_id": "MDQ6VXNlcjg2NDA4MDE5",
"avatar_url": "https://avatars.githubusercontent.com/u/86408019?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ArturoAmorQ",
"html_url": "https://github.com/ArturoAmorQ",
"followers_url": "https://api.github.com/users/ArturoAmorQ/followers",
"following_url": "https://api.github.com/users/ArturoAmorQ/following{/other_user}",
"gists_url": "https://api.github.com/users/ArturoAmorQ/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ArturoAmorQ/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ArturoAmorQ/subscriptions",
"organizations_url": "https://api.github.com/users/ArturoAmorQ/orgs",
"repos_url": "https://api.github.com/users/ArturoAmorQ/repos",
"events_url": "https://api.github.com/users/ArturoAmorQ/events{/privacy}",
"received_events_url": "https://api.github.com/users/ArturoAmorQ/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33400/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33400/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33400",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33400",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33400.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33400.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33399 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33399/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33399/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33399/events | https://github.com/scikit-learn/scikit-learn/pull/33399 | 3,983,950,517 | PR_kwDOAAzd1s7F8f-l | 33,399 | ENH: Display fitted attributes in HTML representation | {
"login": "DeaMariaLeon",
"id": 11835246,
"node_id": "MDQ6VXNlcjExODM1MjQ2",
"avatar_url": "https://avatars.githubusercontent.com/u/11835246?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/DeaMariaLeon",
"html_url": "https://github.com/DeaMariaLeon",
"followers_url": "https://api.github.com/users/DeaMariaLeon/followers",
"following_url": "https://api.github.com/users/DeaMariaLeon/following{/other_user}",
"gists_url": "https://api.github.com/users/DeaMariaLeon/gists{/gist_id}",
"starred_url": "https://api.github.com/users/DeaMariaLeon/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/DeaMariaLeon/subscriptions",
"organizations_url": "https://api.github.com/users/DeaMariaLeon/orgs",
"repos_url": "https://api.github.com/users/DeaMariaLeon/repos",
"events_url": "https://api.github.com/users/DeaMariaLeon/events{/privacy}",
"received_events_url": "https://api.github.com/users/DeaMariaLeon/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815436184,
"node_id": "MDU6TGFiZWwxODE1NDM2MTg0",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:compose",
"name": "module:compose",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 1815453311,
"node_id": "MDU6TGFiZWwxOD... | open | false | [] | null | 15 | 2026-02-24T14:12:25 | 2026-03-06T10:24:14 | null | null | MEMBER | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Supersedes https://github.com/scikit-learn/scikit-learn/pull/31442
See https://github.com/scikit-learn/scikit-learn/pull/31442#issuecomment-3951044495
Towards https://github.com/scikit-learn/scikit-learn/issues/26595
Friendly ping to @jeremiedbb @antoinebaker
#### What does this implement/fix? Explain your changes.
I put the information in 3 columns.
I was going to (try) to modify or remove `ndarray` or try to show it "prettier" but I'm actually not sure if this is what you meant.
<img width="851" height="748" alt="Screenshot 2026-02-24 at 15 09 16" src="https://github.com/user-attachments/assets/e050b4e8-7a8e-42fe-922c-5d41043c8353" />
#### AI
usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [x] Test/benchmark generation -> checked for vulnerabilities on `estimator.py` - EDIT: I didn't add code related to this check for the moment.
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33399/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33399/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33399",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33399",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33399.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33399.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33398 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33398/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33398/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33398/events | https://github.com/scikit-learn/scikit-learn/issues/33398 | 3,983,581,915 | I_kwDOAAzd1s7tcKLb | 33,398 | RFC Collecting "AI infrastructure" for work on scikit-learn | {
"login": "betatim",
"id": 1448859,
"node_id": "MDQ6VXNlcjE0NDg4NTk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1448859?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/betatim",
"html_url": "https://github.com/betatim",
"followers_url": "https://api.github.com/users/betatim/followers",
"following_url": "https://api.github.com/users/betatim/following{/other_user}",
"gists_url": "https://api.github.com/users/betatim/gists{/gist_id}",
"starred_url": "https://api.github.com/users/betatim/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/betatim/subscriptions",
"organizations_url": "https://api.github.com/users/betatim/orgs",
"repos_url": "https://api.github.com/users/betatim/repos",
"events_url": "https://api.github.com/users/betatim/events{/privacy}",
"received_events_url": "https://api.github.com/users/betatim/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1784583753,
"node_id": "MDU6TGFiZWwxNzg0NTgzNzUz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/RFC",
"name": "RFC",
"color": "7fe299",
"default": false,
"description": ""
}
] | open | false | [] | null | 6 | 2026-02-24T12:54:58 | 2026-03-02T16:53:58 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | This issue is the result of a discussion at the monthly maintainer meeting. The goal is to collect ideas and practices from people who use AI tools to work on large projects like scikit-learn. I will call the collection of these ideas and practices "infrastructure" for this issue.
Once we have an overview of ideas and practices we will consider what to do next. One possibility is to create a separate repository that contains the "infrastructure" distilled from the ideas and practices in this issue. People can then checkout that repo "on top" of their scikit-learn checkout and benefit from the content for their AI workflows. Another option is that we add the "infrastructure" directly to the main repository. There are pros and cons to either of these options. There might even be more options, or a compromise.
---
Here are some of my inputs to the collection:
* https://jacobtomlinson.dev/posts/2026/guiding-your-contributors-agents-to-better-behaviours/ a blog post by @jacobtomlinson that talks about exactly this topic. It proposes adding skills to your repository the encode behaviour/limits that you want to be respected. For example, do not work on issues that are missing a "ready" label.
* "design documents" - When working with Cursor to convert an estimator to the array API I found that it got much better after spending some time creating ["design documents" like these](https://gist.github.com/betatim/300d890a8f8bfa2d61ed0f5fdc1f8aa4). I asked cursor to look at existing "array API'ified" estimators, the testing infra, etc and create documents that explain how things work. I then went over them quite carefully to see what it had found. Some of it was pretty good, some of it seemed made up (so I fixed it). In my `AGENTS..md` I refer to `agents/designs/*.md` as a place where it can find relevant information that it should read. I remind myself to ask the agent to keep these up to date with new things it learnt.
* I ask it to make plans and store these in `agents/plans/` with a `current.md` that gives an overview of current and future work. This works semi well, Cursor seems to insist on storing plans somewhere in `~/.cursor/` instead.
* AGENTS.md - I use a variation of https://gist.github.com/betatim/4b205de0b762f8c7865ea2e0ea5c65b4. I recently noticed that for a project the whole section on core principles was missing, but I was sure that it was present (aka I could not really tell the difference in behaviour compared to another project where it was present).
* https://agents.md/ - many examples of `AGENTS.md` files
* https://www.humanlayer.dev/blog/writing-a-good-claude-md opinions on `AGENTS.md` files
My take away/thoughts: a fairly simple `AGENTS.md` seems to be good, it should not contain too much information because that means the fraction of it that is relevant to a task is lower, which "the internet says" reduces the chances of it being followed. Instead it should refer to other documents and state when to read them. While I haven't tried them skills seem to be a good way to help AI users to do the right thing in a repo (eg a "select an issue to work on" skill that encodes the label rules of the repo). There seems to be no need to explain how the scikit-learn build system works, it just works.
What do you think?
cc @AnneBeyer @adrinjalali @ogrisel | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33398/reactions",
"total_count": 4,
"+1": 4,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33398/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33397 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33397/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33397/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33397/events | https://github.com/scikit-learn/scikit-learn/pull/33397 | 3,982,846,627 | PR_kwDOAAzd1s7F4wH0 | 33,397 | CI Fix the doc build (deprecation warning) | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
},
{
"id": 3730476391,
"node_id": "LA_kwDOAAzd1s7eWo1... | closed | false | [] | null | 3 | 2026-02-24T10:33:10 | 2026-02-24T11:02:51 | 2026-02-24T10:43:42 | null | MEMBER | null | null | null | null | See https://github.com/scikit-learn/scikit-learn/pull/30508#issuecomment-3950312489 | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33397/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33397/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33397",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33397",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33397.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33397.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33396 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33396/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33396/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33396/events | https://github.com/scikit-learn/scikit-learn/pull/33396 | 3,982,843,456 | PR_kwDOAAzd1s7F4vgF | 33,396 | DOC Fix deprecated kwargs in `PrecisionRecallDisplay` in example | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
},
{
"id": 3730476391,
"node_id": "LA_kwDOAAzd1s7eW... | closed | false | [] | null | 4 | 2026-02-24T10:32:36 | 2026-02-24T20:28:01 | 2026-02-24T12:17:41 | null | MEMBER | null | null | null | null | #### Reference Issues/PRs
https://github.com/scikit-learn/scikit-learn/pull/30508#issuecomment-3950312489
#### What does this implement/fix? Explain your changes.
Missed in #30508
Uses the new `curve_kwargs` instead of passing plot kwargs as **kwargs, to avoid deprecation warning.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33396/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33396/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33396",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33396",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33396.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33396.patch",
"merged_at": "2026-02-24T12:17:41"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33395 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33395/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33395/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33395/events | https://github.com/scikit-learn/scikit-learn/pull/33395 | 3,982,702,506 | PR_kwDOAAzd1s7F4TVk | 33,395 | FIX SimpleImputer.inverse_transform column order with empty features | {
"login": "worksbyfriday",
"id": 262019507,
"node_id": "U_kgDOD54Zsw",
"avatar_url": "https://avatars.githubusercontent.com/u/262019507?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/worksbyfriday",
"html_url": "https://github.com/worksbyfriday",
"followers_url": "https://api.github.com/users/worksbyfriday/followers",
"following_url": "https://api.github.com/users/worksbyfriday/following{/other_user}",
"gists_url": "https://api.github.com/users/worksbyfriday/gists{/gist_id}",
"starred_url": "https://api.github.com/users/worksbyfriday/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/worksbyfriday/subscriptions",
"organizations_url": "https://api.github.com/users/worksbyfriday/orgs",
"repos_url": "https://api.github.com/users/worksbyfriday/repos",
"events_url": "https://api.github.com/users/worksbyfriday/events{/privacy}",
"received_events_url": "https://api.github.com/users/worksbyfriday/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815443910,
"node_id": "MDU6TGFiZWwxODE1NDQzOTEw",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:impute",
"name": "module:impute",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-02-24T10:06:03 | 2026-02-24T10:29:30 | 2026-02-24T10:29:30 | null | NONE | null | null | null | null | ## Reference Issues/PRs
Fixes #27012
## What does this implement/fix? Explain your changes.
`SimpleImputer.inverse_transform` produces incorrect results when `keep_empty_features=False` (the default) and some columns were all-missing at fit time. Those columns are dropped during `transform`, but `inverse_transform` fails to re-insert them, causing the remaining columns to shift left.
**Root cause**: The existing code used `np.all(X_original[:, original_idx])` to decide whether a column should be skipped during reconstruction. This checks the *test data* indicator values, not whether the column was dropped at fit time.
**Fix**: Use `_get_mask(self.statistics_, np.nan)` to identify which features had no valid statistics at fit time (and were therefore dropped during `transform`). These columns are filled with `missing_values` during reconstruction, preserving correct column alignment.
**Example**:
```python
import numpy as np
from sklearn.impute import SimpleImputer
X_fit = np.array([[np.nan, 2.0, 3.0],
[np.nan, 2.0, 3.0]])
imputer = SimpleImputer(add_indicator=True)
imputer.fit(X_fit)
X_test = np.array([[1.0, 2.0, 3.0],
[1.0, 2.0, 3.0]])
X_trans = imputer.transform(X_test)
# Before fix: columns shifted left, incorrect
# After fix: dropped column correctly re-inserted as NaN
X_inv = imputer.inverse_transform(X_trans)
# array([[nan, 2., 3.],
# [nan, 2., 3.]])
```
Includes a regression test. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33395/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33395/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33395",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33395",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33395.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33395.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33394 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33394/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33394/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33394/events | https://github.com/scikit-learn/scikit-learn/pull/33394 | 3,982,691,835 | PR_kwDOAAzd1s7F4RFj | 33,394 | [Callbacks] TST Add tests to check hook calls | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 3730476391,
"node_id": "... | closed | false | [] | null | 5 | 2026-02-24T10:03:45 | 2026-03-06T09:25:15 | 2026-03-06T09:25:06 | null | MEMBER | null | null | null | null | Towards https://github.com/scikit-learn/scikit-learn/issues/33324
Enhanced a testing callback to record the hook calls and added tests to check the count of callback hooks.
A next step could be to check that the hook calls matches the context tree but let's keep this PR simple.
ping @FrancoisPgm @StefanieSenger | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33394/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33394/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33394",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33394",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33394.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33394.patch",
"merged_at": "2026-03-06T09:25:05"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33393 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33393/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33393/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33393/events | https://github.com/scikit-learn/scikit-learn/pull/33393 | 3,982,634,766 | PR_kwDOAAzd1s7F4E3h | 33,393 | FIX PrecisionRecallDisplay chance level wrong with non-NumPy y_true | {
"login": "worksbyfriday",
"id": 262019507,
"node_id": "U_kgDOD54Zsw",
"avatar_url": "https://avatars.githubusercontent.com/u/262019507?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/worksbyfriday",
"html_url": "https://github.com/worksbyfriday",
"followers_url": "https://api.github.com/users/worksbyfriday/followers",
"following_url": "https://api.github.com/users/worksbyfriday/following{/other_user}",
"gists_url": "https://api.github.com/users/worksbyfriday/gists{/gist_id}",
"starred_url": "https://api.github.com/users/worksbyfriday/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/worksbyfriday/subscriptions",
"organizations_url": "https://api.github.com/users/worksbyfriday/orgs",
"repos_url": "https://api.github.com/users/worksbyfriday/repos",
"events_url": "https://api.github.com/users/worksbyfriday/events{/privacy}",
"received_events_url": "https://api.github.com/users/worksbyfriday/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815445423,
"node_id": "MDU6TGFiZWwxODE1NDQ1NDIz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:metrics",
"name": "module:metrics",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-02-24T09:51:57 | 2026-02-24T10:08:53 | 2026-02-24T10:08:29 | null | NONE | null | null | null | null | Fixes #33342.
## What does this PR do?
`PrecisionRecallDisplay.from_predictions` computed `prevalence_pos_label` using `Counter(y_true)`, which fails for non-NumPy array-like inputs (e.g. PyTorch tensors). `Counter` creates keys that are tensor elements, so looking up `class_count[pos_label]` (where `pos_label` is a Python int) returns 0 — drawing the chance level line at y=0 instead of the true prevalence.
## Fix
Replace the `Counter` approach with `np.asarray` + `np.count_nonzero`, matching the pattern already used in `from_cv_results` (line 903-904). This handles all array-like inputs uniformly.
```python
# Before
class_count = Counter(y_true)
prevalence_pos_label = class_count[pos_label] / sum(class_count.values())
# After
y_true_array = np.asarray(y_true)
prevalence_pos_label = np.count_nonzero(y_true_array == pos_label) / y_true_array.shape[0]
```
The unused `from collections import Counter` import is also removed.
## Test
Added regression test that verifies the chance level line is drawn at the correct prevalence (0.4 for 2/5 positive labels). | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33393/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33393/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33393",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33393",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33393.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33393.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33392 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33392/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33392/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33392/events | https://github.com/scikit-learn/scikit-learn/pull/33392 | 3,982,538,890 | PR_kwDOAAzd1s7F3wiT | 33,392 | FIX check_array TypeError with pandas StringDType | {
"login": "worksbyfriday",
"id": 262019507,
"node_id": "U_kgDOD54Zsw",
"avatar_url": "https://avatars.githubusercontent.com/u/262019507?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/worksbyfriday",
"html_url": "https://github.com/worksbyfriday",
"followers_url": "https://api.github.com/users/worksbyfriday/followers",
"following_url": "https://api.github.com/users/worksbyfriday/following{/other_user}",
"gists_url": "https://api.github.com/users/worksbyfriday/gists{/gist_id}",
"starred_url": "https://api.github.com/users/worksbyfriday/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/worksbyfriday/subscriptions",
"organizations_url": "https://api.github.com/users/worksbyfriday/orgs",
"repos_url": "https://api.github.com/users/worksbyfriday/repos",
"events_url": "https://api.github.com/users/worksbyfriday/events{/privacy}",
"received_events_url": "https://api.github.com/users/worksbyfriday/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815453311,
"node_id": "MDU6TGFiZWwxODE1NDUzMzEx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:utils",
"name": "module:utils",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 0 | 2026-02-24T09:33:07 | 2026-02-24T10:29:44 | 2026-02-24T10:29:44 | null | NONE | null | null | null | null | #### Reference Issues/PRs
Fixes #33383
#### What does this implement/fix? Explain your changes.
Since Pandas 3, `StringDType` is no longer a valid NumPy dtype. `check_array()` unconditionally calls `np.dtype(dtype)` to normalize the dtype parameter for downstream `xp.isdtype()` checks. This raises `TypeError` for non-NumPy-compatible dtypes like pandas `StringDType`.
The fix wraps the `np.dtype()` call in a try/except and falls back to `np.dtype("object")` when the conversion fails. This is correct because:
- String data in NumPy is represented as `object` dtype
- `object` dtype is not integral or floating, so it takes the correct code path through validation (the `else` branch at line 1027)
- `_asarray_with_order` handles the actual array conversion
#### Any other comments?
A changelog entry will be added once the PR number is assigned. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33392/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33392/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33392",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33392",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33392.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33392.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33391 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33391/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33391/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33391/events | https://github.com/scikit-learn/scikit-learn/pull/33391 | 3,982,498,272 | PR_kwDOAAzd1s7F3n5N | 33,391 | FIX LabelBinarizer overflow with unsigned integer dtypes | {
"login": "worksbyfriday",
"id": 262019507,
"node_id": "U_kgDOD54Zsw",
"avatar_url": "https://avatars.githubusercontent.com/u/262019507?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/worksbyfriday",
"html_url": "https://github.com/worksbyfriday",
"followers_url": "https://api.github.com/users/worksbyfriday/followers",
"following_url": "https://api.github.com/users/worksbyfriday/following{/other_user}",
"gists_url": "https://api.github.com/users/worksbyfriday/gists{/gist_id}",
"starred_url": "https://api.github.com/users/worksbyfriday/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/worksbyfriday/subscriptions",
"organizations_url": "https://api.github.com/users/worksbyfriday/orgs",
"repos_url": "https://api.github.com/users/worksbyfriday/repos",
"events_url": "https://api.github.com/users/worksbyfriday/events{/privacy}",
"received_events_url": "https://api.github.com/users/worksbyfriday/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815448196,
"node_id": "MDU6TGFiZWwxODE1NDQ4MTk2",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:preprocessing",
"name": "module:preprocessing",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_id": "MD... | closed | false | [] | null | 2 | 2026-02-24T09:23:56 | 2026-02-24T10:30:13 | 2026-02-24T10:30:02 | null | NONE | null | null | null | null | #### Reference Issues/PRs
Fixes #33390
#### What does this implement/fix? Explain your changes.
When `y` has an unsigned integer dtype (e.g. `uint64`) and `neg_label` is negative (e.g. `-1` in `RidgeClassifier`), `label_binarize()` preserves the unsigned dtype as `int_dtype_`. The cast at line 675 (`xp.astype(Y, int_dtype_, copy=False)`) then silently overflows `-1` to the unsigned maximum (e.g. `18446744073709551615` for `uint64`).
The fix changes the dtype check from `"integral"` to `"signed integer"` so unsigned types fall through to `indexing_dtype(xp)`, which returns a platform-appropriate signed integer type.
#### Any other comments?
The same `"integral"` pattern at line 755 (in `_inverse_binarize_thresholding`) is not affected because it only casts boolean comparison results (`> threshold`) which are always 0 or 1.
A changelog entry will be added once the PR number is assigned. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33391/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33391/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33391",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33391",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33391.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33391.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33390 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33390/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33390/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33390/events | https://github.com/scikit-learn/scikit-learn/issues/33390 | 3,982,136,357 | I_kwDOAAzd1s7tWpQl | 33,390 | RidgeClassifier produces incorrect outputs when y has unsigned integer dtype due to LabelBinarizer | {
"login": "inikishev",
"id": 76593873,
"node_id": "MDQ6VXNlcjc2NTkzODcz",
"avatar_url": "https://avatars.githubusercontent.com/u/76593873?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/inikishev",
"html_url": "https://github.com/inikishev",
"followers_url": "https://api.github.com/users/inikishev/followers",
"following_url": "https://api.github.com/users/inikishev/following{/other_user}",
"gists_url": "https://api.github.com/users/inikishev/gists{/gist_id}",
"starred_url": "https://api.github.com/users/inikishev/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/inikishev/subscriptions",
"organizations_url": "https://api.github.com/users/inikishev/orgs",
"repos_url": "https://api.github.com/users/inikishev/repos",
"events_url": "https://api.github.com/users/inikishev/events{/privacy}",
"received_events_url": "https://api.github.com/users/inikishev/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995755,
"node_id": "MDU6TGFiZWwxMjk5NTc1NQ==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Bug",
"name": "Bug",
"color": "e10c02",
"default": false,
"description": null
},
{
"id": 1815448196,
"node_id": "MDU6TGFiZWwxODE1NDQ4MTk2",
"url"... | open | false | [] | null | 2 | 2026-02-24T07:57:59 | 2026-03-03T04:28:06 | null | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the bug and give evidence about its user-facing impact
I had my targets in UInt64 dtype and RidgeClassifier decision_function was producing giant values. This doesn't happen with other sklearn classifiers and can cause silent and hard to find bugs when RidgeClassifier is used.
### Steps/Code to Reproduce
```python
from sklearn.datasets import make_classification
from sklearn.linear_model import RidgeClassifier
X, y = make_classification(10, 5)
ridge_uint = RidgeClassifier().fit(X, y.astype(np.uint64))
print("UInt64", ridge_uint.decision_function(X)) # large values like 2e+19
ridge_uint = RidgeClassifier().fit(X, y.astype(np.uint32))
print("\nUInt32", ridge_uint.decision_function(X)) # large values like 5e+9
ridge_uint = RidgeClassifier().fit(X, y.astype(np.uint16))
print("\nUInt16", ridge_uint.decision_function(X)) # large values like 8000
ridge_int = RidgeClassifier().fit(X, y.astype(np.int64))
print("\nInt64", ridge_int.decision_function(X)) # normal values
ridge_int = RidgeClassifier().fit(X, y.astype(np.int32))
print("\nInt32", ridge_int.decision_function(X)) # normal values
```
### Expected Results
All dtypes should produce the same results, as in other estimators. If Ridge is not supposed to support UInt, it should raise an exception.
### Actual Results
```
UInt64 [ 1.43620874e+19 1.56493050e+19 2.85214394e+18 1.93398124e+18
-3.17729709e+18 2.12387212e+19 1.52834815e+18 2.45260136e+18
1.73422482e+19 1.80515809e+19]
UInt32 [ 3.34393406e+09 3.64363775e+09 6.64066510e+08 4.50290097e+08
-7.39772126e+08 4.94502514e+09 3.55846285e+08 5.71040755e+08
4.03780681e+09 4.20296121e+09]
UInt16 [ 51023.82734723 55596.80310557 10133.54156372 6871.67191832
-11286.68236037 75453.79278707 5430.61768799 8714.124497
61611.1620848 64131.14136866]
Int64 [-0.55714064 -0.69670104 0.69076993 0.79031733 1.34448324 -1.30270677
0.83429616 0.73408843 -0.88025032 -0.95715633]
Int32 [-0.55714064 -0.69670104 0.69076993 0.79031733 1.34448324 -1.30270677
0.83429616 0.73408843 -0.88025032 -0.95715633]
```
### Versions
```shell
System:
python: 3.13.11 | packaged by Anaconda, Inc. | (main, Dec 10 2025, 21:28:48) [GCC 14.3.0]
executable: /var/mnt/issd/dev/miniconda3/envs/pytorch313/bin/python
machine: Linux-6.17.7-ba25.fc43.x86_64-x86_64-with-glibc2.42
Python dependencies:
sklearn: 1.8.0
pip: 25.3
setuptools: 80.10.1
numpy: 2.4.2
scipy: 1.17.0
Cython: None
pandas: 3.0.1
matplotlib: 3.10.8
joblib: 1.5.3
threadpoolctl: 3.6.0
Built with OpenMP: True
threadpoolctl info:
user_api: blas
internal_api: openblas
num_threads: 8
prefix: libscipy_openblas
filepath: /var/mnt/issd/dev/miniconda3/envs/pytorch313/lib/python3.13/site-packages/numpy.libs/libscipy_openblas64_-096271d3.so
version: 0.3.31.dev
threading_layer: pthreads
architecture: Haswell
user_api: blas
internal_api: openblas
num_threads: 8
prefix: libscipy_openblas
filepath: /var/mnt/issd/dev/miniconda3/envs/pytorch313/lib/python3.13/site-packages/scipy.libs/libscipy_openblas-6cdc3b4a.so
version: 0.3.30
threading_layer: pthreads
architecture: Haswell
user_api: openmp
internal_api: openmp
num_threads: 8
prefix: libgomp
filepath: /var/mnt/issd/dev/miniconda3/envs/pytorch313/lib/python3.13/site-packages/scikit_learn.libs/libgomp-e985bcbb.so.1.0.0
version: None
user_api: openmp
internal_api: openmp
num_threads: 4
prefix: libgomp
filepath: /var/mnt/issd/dev/miniconda3/envs/pytorch313/lib/python3.13/site-packages/torch/lib/libgomp.so.1
version: None
```
### Interest in fixing the bug
this happens in `_RidgeClassifierMixin` in `_prepare_data` method, specifically this line produces incorrect outputs:
```
Y = self._label_binarizer.fit_transform(y)
```
`LabelBinarizer` is initialized with ``pos_label=1, neg_label=-1``, and negative label becomes a large positive number input dtype is unsigned.
In `LabelBinarizer` the overflow happens on line 676, because Y has -1 and is converted to unsigned integer:
```py
Y = xp.astype(Y, int_dtype_, copy=False)
```
```
Y = array([[ 1, -1],
[-1, 1],
[ 1, -1],
[ 1, -1],
[-1, 1],
[-1, 1],
[ 1, -1],
[-1, 1],
[-1, 1],
[ 1, -1]])
int_dtype_ = dtype('uint64')
```
So perhaps a fix or an exception could be added to `LabelBinarizer` | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33390/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33390/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33389 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33389/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33389/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33389/events | https://github.com/scikit-learn/scikit-learn/pull/33389 | 3,981,722,141 | PR_kwDOAAzd1s7F1DgT | 33,389 | Add regression test for LinearSVR thread-safety (liblinear) | {
"login": "GeorgeST1302",
"id": 87616496,
"node_id": "MDQ6VXNlcjg3NjE2NDk2",
"avatar_url": "https://avatars.githubusercontent.com/u/87616496?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/GeorgeST1302",
"html_url": "https://github.com/GeorgeST1302",
"followers_url": "https://api.github.com/users/GeorgeST1302/followers",
"following_url": "https://api.github.com/users/GeorgeST1302/following{/other_user}",
"gists_url": "https://api.github.com/users/GeorgeST1302/gists{/gist_id}",
"starred_url": "https://api.github.com/users/GeorgeST1302/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/GeorgeST1302/subscriptions",
"organizations_url": "https://api.github.com/users/GeorgeST1302/orgs",
"repos_url": "https://api.github.com/users/GeorgeST1302/repos",
"events_url": "https://api.github.com/users/GeorgeST1302/events{/privacy}",
"received_events_url": "https://api.github.com/users/GeorgeST1302/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815448979,
"node_id": "MDU6TGFiZWwxODE1NDQ4OTc5",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:svm",
"name": "module:svm",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-02-24T06:04:34 | 2026-03-03T11:39:26 | 2026-03-03T11:39:26 | null | NONE | null | null | null | null | Refs https://github.com/scikit-learn/scikit-learn/issues/31883
Adds an xfail regression test showing that fitting multiple LinearSVR instances concurrently with joblib threading can yield non-deterministic coefficients.
[x] Code generation
| {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33389/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33389/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33389",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33389",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33389.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33389.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33388 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33388/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33388/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33388/events | https://github.com/scikit-learn/scikit-learn/pull/33388 | 3,981,456,031 | PR_kwDOAAzd1s7F0Lfb | 33,388 | Deprecate probA_ and probB_ in SVC and NuSVC | {
"login": "snath-xoc",
"id": 51656807,
"node_id": "MDQ6VXNlcjUxNjU2ODA3",
"avatar_url": "https://avatars.githubusercontent.com/u/51656807?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/snath-xoc",
"html_url": "https://github.com/snath-xoc",
"followers_url": "https://api.github.com/users/snath-xoc/followers",
"following_url": "https://api.github.com/users/snath-xoc/following{/other_user}",
"gists_url": "https://api.github.com/users/snath-xoc/gists{/gist_id}",
"starred_url": "https://api.github.com/users/snath-xoc/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/snath-xoc/subscriptions",
"organizations_url": "https://api.github.com/users/snath-xoc/orgs",
"repos_url": "https://api.github.com/users/snath-xoc/repos",
"events_url": "https://api.github.com/users/snath-xoc/events{/privacy}",
"received_events_url": "https://api.github.com/users/snath-xoc/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815448979,
"node_id": "MDU6TGFiZWwxODE1NDQ4OTc5",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:svm",
"name": "module:svm",
"color": "151e91",
"default": false,
"description": ""
}
] | open | false | [] | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/milestones/69",
"html_url": "https://github.com/scikit-learn/scikit-learn/milestone/69",
"labels_url": "https://api.github.com/repos/scikit-learn/scikit-learn/milestones/69/labels",
"id": 14107987,
"node_id": "MI_kwDOAAzd1s4A10VT",
"number": 69,
"title": "1.9",
"description": "Planning for the release of 1.9.0: - April 31, 2026: Release candidate 1.9.0.dev0 - May, 18 2026: Release 1.9.0",
"creator": {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
},
"open_issues": 12,
"closed_issues": 6,
"state": "open",
"created_at": "2025-11-12T10:28:02",
"updated_at": "2026-02-27T04:58:45",
"due_on": "2026-05-18T00:00:00",
"closed_at": null
} | 1 | 2026-02-24T04:33:17 | 2026-02-26T16:30:45 | null | null | CONTRIBUTOR | null | null | null | null | #### Reference Issues/PRs
Following up from the deprecation of the `probability=True` in SVC and NuSVC in PR #32050 we now also deprecate the probA_ and probB_ properties. Platt scaling results for probA and probB are still returned in the BaseLibSVM class but unused
#### What does this implement/fix? Explain your changes.
Add deprecation decorator to the probA_ and probB_ properties in BaseSVC
#### Any other comments?
I was getting a prop-decorator error so edited the pyproject.toml Not sure if this has been seen before, perhaps I need to update my mypy? Otherwise I was wondering if the following definitions should also be deprecated so that it is cleaner @jeremiedbb?
- [ ] _check_proba
- [ ] predict_proba
- [ ] predict_log_proba
- [ ] _dense_predict_proba
- [ ] _sparse_predict_proba
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33388/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33388/timeline | null | null | null | true | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33388",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33388",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33388.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33388.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33387 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33387/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33387/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33387/events | https://github.com/scikit-learn/scikit-learn/issues/33387 | 3,981,336,336 | I_kwDOAAzd1s7tTl8Q | 33,387 | ⚠️ CI failed on Unit tests Linux x86-64 pylatest_conda_forge_mkl (last failure: Mar 06, 2026) ⚠️ | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Needs%20Triage",
"name": "Needs Triage",
"color": "FF0000",
"default": false,
"description": "Issue requires triage"
}
] | open | false | [] | null | 3 | 2026-02-24T03:48:09 | 2026-03-06T09:42:17 | null | null | CONTRIBUTOR | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | **CI is still failing on [Unit tests Linux x86-64 pylatest_conda_forge_mkl](https://github.com/scikit-learn/scikit-learn/actions/runs/22747548461/job/65974717202)** (Mar 06, 2026)
- Test Collection Failure
- test_link_inverse_array_api[11-MultinomialLogit-torch-cpu-float32] | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33387/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33387/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33386 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33386/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33386/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33386/events | https://github.com/scikit-learn/scikit-learn/issues/33386 | 3,980,005,153 | I_kwDOAAzd1s7tOg8h | 33,386 | DOC: Clarify max_features default difference between RandomForestClassifier and RandomForestRegressor | {
"login": "jarvy5",
"id": 133603938,
"node_id": "U_kgDOB_aiYg",
"avatar_url": "https://avatars.githubusercontent.com/u/133603938?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jarvy5",
"html_url": "https://github.com/jarvy5",
"followers_url": "https://api.github.com/users/jarvy5/followers",
"following_url": "https://api.github.com/users/jarvy5/following{/other_user}",
"gists_url": "https://api.github.com/users/jarvy5/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jarvy5/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jarvy5/subscriptions",
"organizations_url": "https://api.github.com/users/jarvy5/orgs",
"repos_url": "https://api.github.com/users/jarvy5/repos",
"events_url": "https://api.github.com/users/jarvy5/events{/privacy}",
"received_events_url": "https://api.github.com/users/jarvy5/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995770,
"node_id": "MDU6TGFiZWwxMjk5NTc3MA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Documentation",
"name": "Documentation",
"color": "0b02e1",
"default": false,
"description": null
}
] | open | false | [] | null | 4 | 2026-02-23T20:23:27 | 2026-03-05T17:23:38 | null | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the issue linked to the documentation
While building a Random Forest model for a course project, I found myself confused by the different default values for max_features in classifiers vs regressors and had to search externally to understand why.
In the ensemble documentation (https://scikit-learn.org/stable/modules/ensemble.html),
the `max_features` parameter is mentioned with different defaults for classifiers
(`"sqrt"`) vs regressors (`1.0`), but there is no explanation of why these defaults
differ or how changing this parameter affects model performance in practice.
As a student learning to use Random Forests, I found this confusing when building
my first models. The docs state that `max_features` affects the bias variance tradeoff
but do not clarify the direction of this effect or provide a concrete example.
### Suggest a potential alternative/fix
## Suggested improvement
1. Add a brief explanation of why the defaults differ between classifier and regressor
(the sqrt heuristic comes from Breiman's original paper for classification; using
all features for regression is standard practice to reduce bias)
2. Clarify that lower `max_features` → more randomness between trees → lower variance
but potentially higher bias
3. Add a short code example showing how `max_features` affects out-of-bag score or
test accuracy so users can see the tradeoff empirically
## References
- Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5–32.
- Current docs: [https://scikit-learn.org/stable/modules/ensemble.html#forest]
(https://scikit-learn.org/stable/modules/ensemble.html#forest) | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33386/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33386/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33385 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33385/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33385/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33385/events | https://github.com/scikit-learn/scikit-learn/pull/33385 | 3,979,650,521 | PR_kwDOAAzd1s7FuUwg | 33,385 | FIX Remove unsused arg in callback_management_context | {
"login": "FrancoisPgm",
"id": 35327799,
"node_id": "MDQ6VXNlcjM1MzI3Nzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/35327799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FrancoisPgm",
"html_url": "https://github.com/FrancoisPgm",
"followers_url": "https://api.github.com/users/FrancoisPgm/followers",
"following_url": "https://api.github.com/users/FrancoisPgm/following{/other_user}",
"gists_url": "https://api.github.com/users/FrancoisPgm/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FrancoisPgm/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FrancoisPgm/subscriptions",
"organizations_url": "https://api.github.com/users/FrancoisPgm/orgs",
"repos_url": "https://api.github.com/users/FrancoisPgm/repos",
"events_url": "https://api.github.com/users/FrancoisPgm/events{/privacy}",
"received_events_url": "https://api.github.com/users/FrancoisPgm/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNjU5ODYzNzk1",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/No%20Changelog%20Needed",
"name": "No Changelog Needed",
"color": "B0529C",
"default": false,
"description": ""
},
{
"id": 10245205856,
"node_id": ... | closed | false | [] | null | 0 | 2026-02-23T18:51:01 | 2026-02-24T14:20:02 | 2026-02-23T22:13:29 | null | CONTRIBUTOR | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Towards #27676
#### What does this implement/fix? Explain your changes.
Remove the `fit_method_name` argument of the `callback_management_context` context manager since it not used anymore, as the callback context is no more initialized in the context manager.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
ping @jeremiedbb
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33385/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33385/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33385",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33385",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33385.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33385.patch",
"merged_at": "2026-02-23T22:13:29"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33384 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33384/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33384/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33384/events | https://github.com/scikit-learn/scikit-learn/pull/33384 | 3,979,253,072 | PR_kwDOAAzd1s7FtAaU | 33,384 | Improved README with contribution section | {
"login": "VaishnaviPhalle2005",
"id": 182723383,
"node_id": "U_kgDOCuQjNw",
"avatar_url": "https://avatars.githubusercontent.com/u/182723383?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/VaishnaviPhalle2005",
"html_url": "https://github.com/VaishnaviPhalle2005",
"followers_url": "https://api.github.com/users/VaishnaviPhalle2005/followers",
"following_url": "https://api.github.com/users/VaishnaviPhalle2005/following{/other_user}",
"gists_url": "https://api.github.com/users/VaishnaviPhalle2005/gists{/gist_id}",
"starred_url": "https://api.github.com/users/VaishnaviPhalle2005/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/VaishnaviPhalle2005/subscriptions",
"organizations_url": "https://api.github.com/users/VaishnaviPhalle2005/orgs",
"repos_url": "https://api.github.com/users/VaishnaviPhalle2005/repos",
"events_url": "https://api.github.com/users/VaishnaviPhalle2005/events{/privacy}",
"received_events_url": "https://api.github.com/users/VaishnaviPhalle2005/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 2399338178,
"node_id": "MDU6TGFiZWwyMzk5MzM4MTc4",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/spam",
"name": "spam",
"color": "b60205",
"default": false,
"description": "spam PR and issues"
}
] | closed | false | [] | null | 0 | 2026-02-23T17:12:18 | 2026-02-24T00:21:06 | 2026-02-24T00:21:06 | null | NONE | null | null | null | null | <!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33384/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33384/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33384",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33384",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33384.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33384.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33383 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33383/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33383/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33383/events | https://github.com/scikit-learn/scikit-learn/issues/33383 | 3,979,229,163 | I_kwDOAAzd1s7tLjfr | 33,383 | `check_column_or_1d` doesn't accept `StringDType` as `dtype` parameter since Pandas 3 | {
"login": "folmos-at-orange",
"id": 92923444,
"node_id": "U_kgDOBYnmNA",
"avatar_url": "https://avatars.githubusercontent.com/u/92923444?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/folmos-at-orange",
"html_url": "https://github.com/folmos-at-orange",
"followers_url": "https://api.github.com/users/folmos-at-orange/followers",
"following_url": "https://api.github.com/users/folmos-at-orange/following{/other_user}",
"gists_url": "https://api.github.com/users/folmos-at-orange/gists{/gist_id}",
"starred_url": "https://api.github.com/users/folmos-at-orange/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/folmos-at-orange/subscriptions",
"organizations_url": "https://api.github.com/users/folmos-at-orange/orgs",
"repos_url": "https://api.github.com/users/folmos-at-orange/repos",
"events_url": "https://api.github.com/users/folmos-at-orange/events{/privacy}",
"received_events_url": "https://api.github.com/users/folmos-at-orange/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 12995755,
"node_id": "MDU6TGFiZWwxMjk5NTc1NQ==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Bug",
"name": "Bug",
"color": "e10c02",
"default": false,
"description": null
},
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "ht... | open | false | [] | null | 2 | 2026-02-23T17:07:10 | 2026-03-05T09:47:28 | null | null | NONE | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | ### Describe the bug and give evidence about its user-facing impact
Since Pandas 3.0, `StringDType` is not anymore a valid Numpy data type. This makes `check_column_or_1d` crash when called with that type as `dtype` parameter.
### Steps/Code to Reproduce
```python
from sklearn.utils.validation import column_or_1d
import pandas as pd
pd.options.future.infer_string = True # So this test also works in the 2.x series also
df = pd.DataFrame({"y": ["a", "b", "a", "c"]})
column_or_1d(df["y"], dtype=df["y"].dtype) # fails
```
### Expected Results
No error is thrown.
### Actual Results
```
TypeError Traceback (most recent call last)
Cell In[1], line 8
4 pd.options.future.infer_string = True # So it works in the 2.x series also
7 df = pd.DataFrame({"y": ["a", "b", "a", "c"]})
----> 8 column_or_1d(df["y"], dtype=df["y"].dtype) # fails
File ~/.venv/lib/python3.14/site-packages/sklearn/utils/validation.py:1405, in column_or_1d(y, dtype, input_name, warn, device)
1362 """Ravel column or 1d numpy array, else raises an error.
1363
1364 Parameters
(...) 1402 array([1, 1])
1403 """
1404 xp, _ = get_namespace(y)
-> 1405 y = check_array(
1406 y,
1407 ensure_2d=False,
1408 dtype=dtype,
1409 input_name=input_name,
1410 ensure_all_finite=False,
1411 ensure_min_samples=0,
1412 )
1414 shape = y.shape
1415 if len(shape) == 1:
File ~/.venv/lib/python3.14/site-packages/sklearn/utils/validation.py:952, in check_array(array, accept_sparse, accept_large_sparse, dtype, order, copy, force_writeable, ensure_all_finite, ensure_non_negative, ensure_2d, allow_nd, ensure_min_samples, ensure_min_features, estimator, input_name)
945 raise ValueError(
946 "ensure_all_finite should be a bool or 'allow-nan'. Got "
947 f"{ensure_all_finite!r} instead."
948 )
950 if dtype is not None and _is_numpy_namespace(xp):
951 # convert to dtype object to conform to Array API to be use `xp.isdtype` later
--> 952 dtype = np.dtype(dtype)
954 estimator_name = _check_estimator_name(estimator)
955 context = " by %s" % estimator_name if estimator is not None else ""
TypeError: Cannot interpret '<StringDtype(storage='python', na_value=nan)>' as a data type
```
### Versions
```shell
System:
python: 3.14.3 (main, Feb 3 2026, 15:32:20) [Clang 17.0.0 (clang-1700.6.3.2)]
executable: ~/.cache/uv/builds-v0/.tmphptoZv/bin/python
machine: macOS-26.3-arm64-arm-64bit-Mach-O
Python dependencies:
sklearn: 1.8.0
pip: None
setuptools: None
numpy: 2.4.2
scipy: 1.17.1
Cython: None
pandas: 3.0.1
matplotlib: None
joblib: 1.5.3
threadpoolctl: 3.6.0
Built with OpenMP: True
threadpoolctl info:
user_api: openmp
internal_api: openmp
num_threads: 10
prefix: libomp
filepath: /<redacted>/.venv/lib/python3.14/site-packages/sklearn/.dylibs/libomp.dylib
version: None
```
### Interest in fixing the bug
Yes | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33383/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33383/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33382 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33382/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33382/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33382/events | https://github.com/scikit-learn/scikit-learn/pull/33382 | 3,978,763,275 | PR_kwDOAAzd1s7FrZdV | 33,382 | MAINT Update error message for liblinear with penalty=None | {
"login": "shivansh023023",
"id": 168457476,
"node_id": "U_kgDOCgp1BA",
"avatar_url": "https://avatars.githubusercontent.com/u/168457476?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shivansh023023",
"html_url": "https://github.com/shivansh023023",
"followers_url": "https://api.github.com/users/shivansh023023/followers",
"following_url": "https://api.github.com/users/shivansh023023/following{/other_user}",
"gists_url": "https://api.github.com/users/shivansh023023/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shivansh023023/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shivansh023023/subscriptions",
"organizations_url": "https://api.github.com/users/shivansh023023/orgs",
"repos_url": "https://api.github.com/users/shivansh023023/repos",
"events_url": "https://api.github.com/users/shivansh023023/events{/privacy}",
"received_events_url": "https://api.github.com/users/shivansh023023/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815444632,
"node_id": "MDU6TGFiZWwxODE1NDQ0NjMy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:linear_model",
"name": "module:linear_model",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 9691732262,
"node_id": "LA_k... | closed | false | [] | null | 4 | 2026-02-23T15:27:27 | 2026-02-23T15:42:30 | 2026-02-23T15:32:40 | null | NONE | null | null | null | null | #### Reference Issues/PRs
This addresses the `# TODO(1.10)` found in `sklearn/linear_model/_logistic.py` regarding the simplification of the error message for the `liblinear` solver when `penalty=None`.
#### What does this implement/fix? Explain your changes.
This PR performs two main updates:
1. **Logic Update**: In `sklearn/linear_model/_logistic.py`, I updated the `ValueError` message in `_check_solver` to remove the redundant reference to `C=np.inf`, as requested by the internal TODO.
2. **Test Update**: In `sklearn/linear_model/tests/test_logistic.py`, I updated `test_check_solver_option` to match the new error string. I also modified the test call to use a slice of the data (`X[:100], y[:100]`) to ensure the test remains binary, avoiding a multiclass validation error that was masking the penalty check.
#### AI usage disclosure
I used AI assistance for:
- [x] Code generation (Used to help identify the string replacement and draft the test fix)
- [x] Test/benchmark generation (Assisted in debugging why the multiclass error was triggered during testing)
- [ ] Documentation (including examples)
- [x] Research and understanding (Used to understand the relationship between C=np.inf and penalty=None in this context)
*Note: All AI-generated suggestions were manually reviewed, implemented, and verified by running the full test suite locally.*
#### Any other comments?
I have verified these changes by running `pytest sklearn/linear_model/tests/test_logistic.py` locally. All 284 tests passed. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33382/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33382/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33382",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33382",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33382.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33382.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33381 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33381/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33381/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33381/events | https://github.com/scikit-learn/scikit-learn/pull/33381 | 3,978,757,893 | PR_kwDOAAzd1s7FrYV4 | 33,381 | Enh 33380 remove fmi sparse | {
"login": "un1u3",
"id": 112570655,
"node_id": "U_kgDOBrWxHw",
"avatar_url": "https://avatars.githubusercontent.com/u/112570655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/un1u3",
"html_url": "https://github.com/un1u3",
"followers_url": "https://api.github.com/users/un1u3/followers",
"following_url": "https://api.github.com/users/un1u3/following{/other_user}",
"gists_url": "https://api.github.com/users/un1u3/gists{/gist_id}",
"starred_url": "https://api.github.com/users/un1u3/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/un1u3/subscriptions",
"organizations_url": "https://api.github.com/users/un1u3/orgs",
"repos_url": "https://api.github.com/users/un1u3/repos",
"events_url": "https://api.github.com/users/un1u3/events{/privacy}",
"received_events_url": "https://api.github.com/users/un1u3/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815445423,
"node_id": "MDU6TGFiZWwxODE1NDQ1NDIz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:metrics",
"name": "module:metrics",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-02-23T15:26:19 | 2026-02-23T16:48:50 | 2026-02-23T16:42:07 | null | CONTRIBUTOR | null | null | null | null | Closes #33380
- [x] Triaged TODO(1.9) items and completed this one.
- [x] Removed deprecated `sparse` parameter from `metrics.fowlkes_mallows_score`.
- [x] Removed deprecated warning/handling path.
- [x] Removed related deprecation test.
- [x] Updated parameter validation/docstring accordingly. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33381/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33381/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33381",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33381",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33381.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33381.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33380 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33380/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33380/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33380/events | https://github.com/scikit-learn/scikit-learn/issues/33380 | 3,978,664,099 | I_kwDOAAzd1s7tJZij | 33,380 | ENH: Remove deprecated sparse parameter from metrics.fowlkes_mallows_score (1.9 TODO) | {
"login": "un1u3",
"id": 112570655,
"node_id": "U_kgDOBrWxHw",
"avatar_url": "https://avatars.githubusercontent.com/u/112570655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/un1u3",
"html_url": "https://github.com/un1u3",
"followers_url": "https://api.github.com/users/un1u3/followers",
"following_url": "https://api.github.com/users/un1u3/following{/other_user}",
"gists_url": "https://api.github.com/users/un1u3/gists{/gist_id}",
"starred_url": "https://api.github.com/users/un1u3/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/un1u3/subscriptions",
"organizations_url": "https://api.github.com/users/un1u3/orgs",
"repos_url": "https://api.github.com/users/un1u3/repos",
"events_url": "https://api.github.com/users/un1u3/events{/privacy}",
"received_events_url": "https://api.github.com/users/un1u3/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Needs%20Triage",
"name": "Needs Triage",
"color": "FF0000",
"default": false,
"description": "Issue requires triage"
}
] | closed | false | [] | null | 0 | 2026-02-23T15:07:13 | 2026-02-23T16:49:54 | 2026-02-23T16:49:54 | null | CONTRIBUTOR | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | The sparse parameter in `metrics.fowlkes_mallows_score` was deprecated in 1.7 and has no effect. It is scheduled for removal in 1.9.
Goal: remove sparse from the function signature, warnings, and tests.
<img width="977" height="215" alt="Image" src="https://github.com/user-attachments/assets/65d18324-fb0b-4006-b044-10f6bdac5428" /> | {
"login": "un1u3",
"id": 112570655,
"node_id": "U_kgDOBrWxHw",
"avatar_url": "https://avatars.githubusercontent.com/u/112570655?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/un1u3",
"html_url": "https://github.com/un1u3",
"followers_url": "https://api.github.com/users/un1u3/followers",
"following_url": "https://api.github.com/users/un1u3/following{/other_user}",
"gists_url": "https://api.github.com/users/un1u3/gists{/gist_id}",
"starred_url": "https://api.github.com/users/un1u3/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/un1u3/subscriptions",
"organizations_url": "https://api.github.com/users/un1u3/orgs",
"repos_url": "https://api.github.com/users/un1u3/repos",
"events_url": "https://api.github.com/users/un1u3/events{/privacy}",
"received_events_url": "https://api.github.com/users/un1u3/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33380/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33380/timeline | null | not_planned | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33379 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33379/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33379/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33379/events | https://github.com/scikit-learn/scikit-learn/pull/33379 | 3,978,447,637 | PR_kwDOAAzd1s7FqWLQ | 33,379 | ENH Add partial_fit method to NearestCentroid for incremental learning,Fix lint and mypy issues | {
"login": "alobroke",
"id": 170533827,
"node_id": "U_kgDOCiojww",
"avatar_url": "https://avatars.githubusercontent.com/u/170533827?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alobroke",
"html_url": "https://github.com/alobroke",
"followers_url": "https://api.github.com/users/alobroke/followers",
"following_url": "https://api.github.com/users/alobroke/following{/other_user}",
"gists_url": "https://api.github.com/users/alobroke/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alobroke/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alobroke/subscriptions",
"organizations_url": "https://api.github.com/users/alobroke/orgs",
"repos_url": "https://api.github.com/users/alobroke/repos",
"events_url": "https://api.github.com/users/alobroke/events{/privacy}",
"received_events_url": "https://api.github.com/users/alobroke/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 9691732262,
"node_id": "LA_kwDOAAzd1s8AAAACQawZJg",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/CI:Linter%20failure",
"name": "CI:Linter failure",
"color": "f42108",
"default": false,
"description": "The linter CI is failing on this PR"
}
] | closed | false | [] | null | 2 | 2026-02-23T14:24:45 | 2026-02-24T01:56:40 | 2026-02-24T01:56:40 | null | NONE | null | null | null | null | ## Summary
This PR adds a `partial_fit` method to the `NearestCentroid` classifier to enable incremental (online) learning.
The implementation updates class centroids using a weighted mean based on the number of samples observed per class, allowing the model to be trained on data batches without retraining from scratch.
## Details
- Maintains `class_count_` to track number of samples per class
- Updates `centroids_` incrementally using weighted averaging
- Supports multiple calls to `partial_fit`
- Handles initialization during first call
- Follows design patterns similar to `GaussianNB.partial_fit`
## Motivation
This enhancement enables `NearestCentroid` to work efficiently in streaming, large-scale, and memory-constrained environments where full retraining is not feasible.
## Testing
- Added unit test to verify incremental centroid updates
- All linting, formatting, and type checks pass locally
## Related Issue
Closes #12952
<!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
#### What does this implement/fix? Explain your changes.
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33379/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33379/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33379",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33379",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33379.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33379.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33378 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33378/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33378/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33378/events | https://github.com/scikit-learn/scikit-learn/pull/33378 | 3,978,374,438 | PR_kwDOAAzd1s7FqGh0 | 33,378 | ENH Add partial_fit method to NearestCentroid | {
"login": "alobroke",
"id": 170533827,
"node_id": "U_kgDOCiojww",
"avatar_url": "https://avatars.githubusercontent.com/u/170533827?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alobroke",
"html_url": "https://github.com/alobroke",
"followers_url": "https://api.github.com/users/alobroke/followers",
"following_url": "https://api.github.com/users/alobroke/following{/other_user}",
"gists_url": "https://api.github.com/users/alobroke/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alobroke/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alobroke/subscriptions",
"organizations_url": "https://api.github.com/users/alobroke/orgs",
"repos_url": "https://api.github.com/users/alobroke/repos",
"events_url": "https://api.github.com/users/alobroke/events{/privacy}",
"received_events_url": "https://api.github.com/users/alobroke/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815447666,
"node_id": "MDU6TGFiZWwxODE1NDQ3NjY2",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:neighbors",
"name": "module:neighbors",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 9691732262,
"node_id": "LA_kwDOAAz... | open | false | [] | null | 1 | 2026-02-23T14:07:41 | 2026-02-23T14:08:31 | null | null | NONE | null | null | null | null | Adds incremental learning support via partial_fit method.
Centroids are updated using weighted mean based on class counts.
Includes tests and documentation. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33378/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33378/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33378",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33378",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33378.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33378.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33377 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33377/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33377/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33377/events | https://github.com/scikit-learn/scikit-learn/pull/33377 | 3,978,114,273 | PR_kwDOAAzd1s7FpP9k | 33,377 | FIX make test more robust by increasing dataset size | {
"login": "AnneBeyer",
"id": 12881170,
"node_id": "MDQ6VXNlcjEyODgxMTcw",
"avatar_url": "https://avatars.githubusercontent.com/u/12881170?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AnneBeyer",
"html_url": "https://github.com/AnneBeyer",
"followers_url": "https://api.github.com/users/AnneBeyer/followers",
"following_url": "https://api.github.com/users/AnneBeyer/following{/other_user}",
"gists_url": "https://api.github.com/users/AnneBeyer/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AnneBeyer/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AnneBeyer/subscriptions",
"organizations_url": "https://api.github.com/users/AnneBeyer/orgs",
"repos_url": "https://api.github.com/users/AnneBeyer/repos",
"events_url": "https://api.github.com/users/AnneBeyer/events{/privacy}",
"received_events_url": "https://api.github.com/users/AnneBeyer/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815435821,
"node_id": "MDU6TGFiZWwxODE1NDM1ODIx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:cluster",
"name": "module:cluster",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2659863795,
"node_id": "MDU6TGFiZWwyNj... | closed | false | [] | null | 2 | 2026-02-23T13:08:48 | 2026-02-24T13:28:21 | 2026-02-24T13:28:21 | null | CONTRIBUTOR | null | null | null | null |
<!--
🙌 Thanks for contributing a pull request!
👀 Please ensure you have taken a look at the contribution guidelines:
https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md
✅ In particular following the pull request checklist will increase the likelihood
of having maintainers review your PR:
https://scikit-learn.org/dev/developers/contributing.html#pull-request-checklist
📋 If your PR is likely to affect users, you will need to add a changelog entry
describing your PR changes, see:
https://github.com/scikit-learn/scikit-learn/blob/main/doc/whats_new/upcoming_changes/README.md
-->
#### Reference Issues/PRs
<!--
Example: Fixes #1234. See also #3456.
Please use keywords (e.g., Fixes) to create link to the issues or pull requests
you resolved, so that they will automatically be closed when your pull request
is merged. See https://github.com/blog/1506-closing-issues-via-pull-requests
-->
Fixes #33365
#### What does this implement/fix? Explain your changes.
Increasing the dataset size from 250 to 300 (as suggested [here](https://github.com/scikit-learn/scikit-learn/issues/33365#issuecomment-3944599860)) fixes the issue for me locally.
Building with `[all random seeds]` here to see if CI is also happy with this fix.
@lesteve
#### AI usage disclosure
<!--
If AI tools were involved in creating this PR, please check all boxes that apply
below and make sure that you adhere to our Automated Contributions Policy:
https://scikit-learn.org/dev/developers/contributing.html#automated-contributions-policy
-->
I used AI assistance for:
- [ ] Code generation (e.g., when writing an implementation or fixing a bug)
- [ ] Test/benchmark generation
- [ ] Documentation (including examples)
- [ ] Research and understanding
#### Any other comments?
<!--
Thank you for your patience. Changes to scikit-learn require careful
attention, but with limited maintainer time, not every contribution can be reviewed
quickly.
For more information and tips on improving your pull request, see:
https://scikit-learn.org/dev/faq.html#why-is-my-pull-request-not-getting-any-attention.
Thanks for contributing!
-->
| {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33377/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33377/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33377",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33377",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33377.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33377.patch",
"merged_at": "2026-02-24T13:28:21"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33376 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33376/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33376/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33376/events | https://github.com/scikit-learn/scikit-learn/pull/33376 | 3,977,313,061 | PR_kwDOAAzd1s7Fmmf_ | 33,376 | fix: honor copy=False for y in mutual_info and clarify docstring | {
"login": "dhruvildarji",
"id": 25696982,
"node_id": "MDQ6VXNlcjI1Njk2OTgy",
"avatar_url": "https://avatars.githubusercontent.com/u/25696982?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dhruvildarji",
"html_url": "https://github.com/dhruvildarji",
"followers_url": "https://api.github.com/users/dhruvildarji/followers",
"following_url": "https://api.github.com/users/dhruvildarji/following{/other_user}",
"gists_url": "https://api.github.com/users/dhruvildarji/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dhruvildarji/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dhruvildarji/subscriptions",
"organizations_url": "https://api.github.com/users/dhruvildarji/orgs",
"repos_url": "https://api.github.com/users/dhruvildarji/repos",
"events_url": "https://api.github.com/users/dhruvildarji/events{/privacy}",
"received_events_url": "https://api.github.com/users/dhruvildarji/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815442108,
"node_id": "MDU6TGFiZWwxODE1NDQyMTA4",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:feature_selection",
"name": "module:feature_selection",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2399338178,
"node_... | closed | false | [] | null | 1 | 2026-02-23T10:00:41 | 2026-03-04T00:24:51 | 2026-03-04T00:24:39 | null | NONE | null | null | null | null | Fixes #28793.
**Problem:** The `copy` parameter docstring for `mutual_info_regression` and `mutual_info_classif` states "If set to False, the initial data will be overwritten", but `y` was never overwritten because `scale(y, with_mean=False)` did not pass `copy=copy` (defaulting to `copy=True`).
**Changes:**
- Pass `copy=copy` to `scale(y, with_mean=False, copy=copy)` in `_estimate_mi` so the parameter is honored for continuous `y` in `mutual_info_regression`
- Clarify the `copy` docstring in both `mutual_info_regression` and `mutual_info_classif` to accurately describe which data is affected | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33376/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33376/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33376",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33376",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33376.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33376.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33374 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33374/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33374/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33374/events | https://github.com/scikit-learn/scikit-learn/pull/33374 | 3,977,076,317 | PR_kwDOAAzd1s7Fl0bn | 33,374 | ENH: add zero_division parameter to ndcg_score | {
"login": "dhruvildarji",
"id": 25696982,
"node_id": "MDQ6VXNlcjI1Njk2OTgy",
"avatar_url": "https://avatars.githubusercontent.com/u/25696982?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dhruvildarji",
"html_url": "https://github.com/dhruvildarji",
"followers_url": "https://api.github.com/users/dhruvildarji/followers",
"following_url": "https://api.github.com/users/dhruvildarji/following{/other_user}",
"gists_url": "https://api.github.com/users/dhruvildarji/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dhruvildarji/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dhruvildarji/subscriptions",
"organizations_url": "https://api.github.com/users/dhruvildarji/orgs",
"repos_url": "https://api.github.com/users/dhruvildarji/repos",
"events_url": "https://api.github.com/users/dhruvildarji/events{/privacy}",
"received_events_url": "https://api.github.com/users/dhruvildarji/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815445423,
"node_id": "MDU6TGFiZWwxODE1NDQ1NDIz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:metrics",
"name": "module:metrics",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-02-23T09:03:17 | 2026-03-04T00:23:54 | 2026-03-04T00:23:53 | null | NONE | null | null | null | null | ## Summary
Fixes #29521.
This PR adds a `zero_division` parameter to `ndcg_score` (and the internal `_ndcg_sample_scores`) to handle the case where all true relevance scores for a sample are zero (IDCG = 0, so NDCG is undefined).
This follows the pattern used by other scikit-learn metrics (e.g., `f1_score`, `precision_score`) and addresses the guidance from @glemaitre in the issue.
### Behavior
| `zero_division` | Behavior when IDCG = 0 |
|---|---|
| `"warn"` (default) | Returns `0` for that sample **and** raises `UndefinedMetricWarning` |
| `0` | Returns `0` silently (previous behavior, now without warning) |
| `1` | Returns `1` silently |
### Changes
- **`sklearn/metrics/_ranking.py`**: Added `zero_division` parameter to both `ndcg_score` and `_ndcg_sample_scores`. Updated `@validate_params` decorator to validate the new parameter using `StrOptions({"warn"})` and `Options(Real, {0, 1})`. Updated docstrings.
- **`sklearn/metrics/tests/test_ranking.py`**: Added `test_ndcg_score_zero_division` covering all three parameter values.
- **`doc/whats_new/upcoming_changes/sklearn.metrics/29521.enhancement.rst`**: Added changelog entry.
### Example
```python
import numpy as np
from sklearn.metrics import ndcg_score
y_true = np.array([[1.0, 0.0, 1.0], [0.0, 0.0, 0.0]])
y_score = np.array([[0.5, 0.2, 0.8], [0.1, 0.3, 0.6]])
# Default "warn": raises UndefinedMetricWarning, zero sample = 0
ndcg_score(y_true, y_score) # UndefinedMetricWarning raised
# Explicit 0: silent, zero sample = 0
ndcg_score(y_true, y_score, zero_division=0)
# Explicit 1: silent, zero sample = 1
ndcg_score(y_true, y_score, zero_division=1)
```
Note: The previous default behavior was to silently set NDCG=0 for samples with all-zero true relevance. The new default `"warn"` preserves that return value but now raises a warning so users are informed that those samples are undefined. This is a minor behavioral change (new warning), consistent with how other metrics handle this case. | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33374/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33374/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33374",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33374",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33374.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33374.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33373 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33373/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33373/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33373/events | https://github.com/scikit-learn/scikit-learn/issues/33373 | 3,976,543,871 | I_kwDOAAzd1s7tBT5_ | 33,373 | RFC When to use `_max_precision_float_dtype` in metrics? | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1784583753,
"node_id": "MDU6TGFiZWwxNzg0NTgzNzUz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/RFC",
"name": "RFC",
"color": "7fe299",
"default": false,
"description": ""
},
{
"id": 5464227604,
"node_id": "LA_kwDOAAzd1s8AAAABRbF_FA",
"url... | open | false | [] | null | 3 | 2026-02-23T06:20:13 | 2026-02-28T00:03:21 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | > [!WARNING]
> This is a request for comment (RFC) and is not open for contribution. Great if you are interested to contribute to scikit-learn :pray:. Please have a look at our [contributing doc](https://scikit-learn.org/dev/developers/contributing.html) and in particular the section [New Contributors](https://scikit-learn.org/dev/developers/contributing.html#new-contributors).
Noticed while looking into #33371 and #33366
There are a few places in array API metrics where we use `_max_precision_float_dtype` but the note does not seem to (adequately) explain why `_max_precision_float_dtype` was used and not just `_find_matching_floating_dtype`. E.g.,
`cohen_kappa_score` added in #32619
https://github.com/scikit-learn/scikit-learn/blob/c4f0dbb3f350ac015368b83ebca607d5d735c13d/sklearn/metrics/_classification.py#L995-L1000
Again in `precision_recall_fscore_support` (#30395):
https://github.com/scikit-learn/scikit-learn/blob/c4f0dbb3f350ac015368b83ebca607d5d735c13d/sklearn/metrics/_classification.py#L2191-L2197
`balanced_accuracy_score` (#32604)
https://github.com/scikit-learn/scikit-learn/blob/c4f0dbb3f350ac015368b83ebca607d5d735c13d/sklearn/metrics/_classification.py#L2959-L2962
(Other examples, not listed)
We also do this in `davies_bouldin_score` (#32693):
https://github.com/scikit-learn/scikit-learn/blob/c4f0dbb3f350ac015368b83ebca607d5d735c13d/sklearn/metrics/cluster/_unsupervised.py#L465-L467
And I did ask why, but the contributor did not know and we never questioned it during review: https://github.com/scikit-learn/scikit-learn/pull/32693#discussion_r2566984547
This may just be my mis-understanding, or maybe we could amend the comment slightly.
cc @ogrisel @betatim @OmarManzoor
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33373/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33373/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33371 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33371/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33371/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33371/events | https://github.com/scikit-learn/scikit-learn/issues/33371 | 3,976,456,179 | I_kwDOAAzd1s7tA-fz | 33,371 | RFC Should metrics output dtype match input? | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1784583753,
"node_id": "MDU6TGFiZWwxNzg0NTgzNzUz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/RFC",
"name": "RFC",
"color": "7fe299",
"default": false,
"description": ""
},
{
"id": 5464227604,
"node_id": "LA_kwDOAAzd1s8AAAABRbF_FA",
"url... | closed | false | [] | null | 4 | 2026-02-23T05:45:13 | 2026-02-27T01:14:28 | 2026-02-27T01:14:28 | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | > [!WARNING]
> This is a request for comment (RFC) and is not open for contribution. Great if you are interested to contribute to scikit-learn :pray:. Please have a look at our [contributing doc](https://scikit-learn.org/dev/developers/contributing.html) and in particular the section [New Contributors](https://scikit-learn.org/dev/developers/contributing.html#new-contributors).
Noticed while writing #33366
There are metrics where we use `_max_precision_float_dtype` during computation but do not update the output to be input dtype. Should the output dtype match input (namely `y_pred`) dtype?
e.g.,
```python
import numpy as np
from sklearn.metrics import det_curve
y_true = np.array([0, 0, 1, 1, 0, 1, 1, 0, 1, 0])
y_score = np.array([0.1, 0.4, 0.35, 0.8, 0.2, 0.7, 0.9, 0.3, 0.6, 0.15], dtype=np.float32)
fpr, fnr, thresholds = det_curve(y_true, y_score)
```
dtype of the outputs are float64.
Edit: more fleshed out pros and cons
Pro
* To be consistent with what happens in estimators
* if one of these upcasting metrics is passed to e.g., CV estimator, the `cv_results` attribute will be the same as input dtype, whereas using that metrics by itself, will give you float64
Con
* Most metric outputs are [python float](https://docs.python.org/3/library/stdtypes.html#numeric-types-int-float-complex), which (I think) are usually 64 anyway
cc @ogrisel @betatim @OmarManzoor | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33371/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33371/timeline | null | completed | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33370 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33370/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33370/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33370/events | https://github.com/scikit-learn/scikit-learn/pull/33370 | 3,976,424,835 | PR_kwDOAAzd1s7FjruY | 33,370 | :lock: :robot: CI Update lock files for main CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 0 | 2026-02-23T05:33:08 | 2026-02-23T09:16:47 | 2026-02-23T09:16:47 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33370/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33370/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33370",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33370",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33370.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33370.patch",
"merged_at": "2026-02-23T09:16:47"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33369 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33369/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33369/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33369/events | https://github.com/scikit-learn/scikit-learn/pull/33369 | 3,976,414,588 | PR_kwDOAAzd1s7Fjpel | 33,369 | :lock: :robot: CI Update lock files for array-api CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 0 | 2026-02-23T05:30:20 | 2026-02-23T09:15:34 | 2026-02-23T09:15:34 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33369/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33369/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33369",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33369",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33369.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33369.patch",
"merged_at": "2026-02-23T09:15:33"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33368 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33368/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33368/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33368/events | https://github.com/scikit-learn/scikit-learn/pull/33368 | 3,976,414,019 | PR_kwDOAAzd1s7FjpWn | 33,368 | :lock: :robot: CI Update lock files for scipy-dev CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 0 | 2026-02-23T05:30:10 | 2026-02-23T09:15:07 | 2026-02-23T09:15:07 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33368/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33368/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33368",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33368",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33368.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33368.patch",
"merged_at": "2026-02-23T09:15:07"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33367 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33367/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33367/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33367/events | https://github.com/scikit-learn/scikit-learn/pull/33367 | 3,976,413,630 | PR_kwDOAAzd1s7FjpRA | 33,367 | :lock: :robot: CI Update lock files for free-threaded CI build(s) :lock: :robot: | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 55523804,
"node_id": "MDU6TGFiZWw1NTUyMzgwNA==",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Build%20/%20CI",
"name": "Build / CI",
"color": "eb6420",
"default": false,
"description": null
}
] | closed | false | [] | null | 0 | 2026-02-23T05:30:04 | 2026-02-23T09:14:48 | 2026-02-23T09:14:47 | null | CONTRIBUTOR | null | null | null | null | Update lock files.
### Note
If the CI tasks fail, create a new branch based on this PR and add the required fixes to that branch. | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33367/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33367/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33367",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33367",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33367.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33367.patch",
"merged_at": "2026-02-23T09:14:47"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33366 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33366/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33366/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33366/events | https://github.com/scikit-learn/scikit-learn/issues/33366 | 3,976,407,565 | I_kwDOAAzd1s7tAyoN | 33,366 | RFC: dtype downcasting and precision in array API | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1784583753,
"node_id": "MDU6TGFiZWwxNzg0NTgzNzUz",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/RFC",
"name": "RFC",
"color": "7fe299",
"default": false,
"description": ""
},
{
"id": 5464227604,
"node_id": "LA_kwDOAAzd1s8AAAABRbF_FA",
"url... | open | false | [] | null | 0 | 2026-02-23T05:27:58 | 2026-02-23T05:29:17 | null | null | MEMBER | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | > [!WARNING]
> This is a request for comment (RFC) and is not open for contribution. Great if you are interested to contribute to scikit-learn :pray:. Please have a look at our [contributing doc](https://scikit-learn.org/dev/developers/contributing.html) and in particular the section [New Contributors](https://scikit-learn.org/dev/developers/contributing.html#new-contributors).
Some metric functions upcast to max float type (via `_max_precision_float_dtype`) during computation; for example `confusion_matrix_at_thresholds` does this because it uses `cumulative_sum`, which suffers from precision degradation due to error accumulation. (Aside, note that we return arrays at max dtype, and do not convert back to input dtype).
Some estimators also upcast during computation, e.g., `RidgeCV`, but we return attribute arrays to input array dtype (AFAICT) so the 'output' dtype is matches input dtype.
If the user is using a device that only supports float 32 (i.e. MPS), it is possible that they suffer from precision problems when using these metrics/estimators. Currently, we inform users of this by adding a "(see [Note on device support for float64](https://scikit-learn.org/dev/modules/array_api.html#device-support-for-float64))" beside that estimator/metric in our docs.
This also affects the mixed array namespace/device inputs situation (wrt metric functions see #31274 and #32755 and wrt fit and predict see #33076). If we are moving arrays to a device where only float32 is supported, precision may be affected for the above estimators/metrics.
How do we remember to document this and make it clear to people reading the code that a downcast is occurring? Options discussed during the last array API meeting (not mutually exclusive):
* Use existing estimator tag system to tag estimators affected
* this excludes metrics. Note that although we do check that the output of metrics with different array namespace inputs matches that of numpy, this will not always pick up on precision problems as it is data dependent
* Add a `dtype` kwarg to `move_to` - this would make it clearer in the code that a potential downcast was occurring (see discussion https://github.com/scikit-learn/scikit-learn/pull/32755#discussion_r2611300838)
* though this would make it more difficult to use the same `move_to` function for both array conversion and 'converting' an estimator, see: https://github.com/scikit-learn/scikit-learn/pull/33076#discussion_r2822481554
@ogrisel @betatim @lesteve I may have missed something or mis-understood something, please update if I have. | null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33366/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33366/timeline | null | null | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33365 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33365/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33365/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33365/events | https://github.com/scikit-learn/scikit-learn/issues/33365 | 3,976,147,490 | I_kwDOAAzd1s7s_zIi | 33,365 | ⚠️ CI failed on Unit tests Linux x86-64 pymin_conda_forge_openblas_ubuntu_2204 (last failure: Feb 23, 2026) ⚠️ | {
"login": "scikit-learn-bot",
"id": 8699527,
"node_id": "MDQ6VXNlcjg2OTk1Mjc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8699527?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/scikit-learn-bot",
"html_url": "https://github.com/scikit-learn-bot",
"followers_url": "https://api.github.com/users/scikit-learn-bot/followers",
"following_url": "https://api.github.com/users/scikit-learn-bot/following{/other_user}",
"gists_url": "https://api.github.com/users/scikit-learn-bot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/scikit-learn-bot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/scikit-learn-bot/subscriptions",
"organizations_url": "https://api.github.com/users/scikit-learn-bot/orgs",
"repos_url": "https://api.github.com/users/scikit-learn-bot/repos",
"events_url": "https://api.github.com/users/scikit-learn-bot/events{/privacy}",
"received_events_url": "https://api.github.com/users/scikit-learn-bot/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 3631152487,
"node_id": "LA_kwDOAAzd1s7Ybv1n",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/Needs%20Triage",
"name": "Needs Triage",
"color": "FF0000",
"default": false,
"description": "Issue requires triage"
}
] | closed | false | [] | null | 3 | 2026-02-23T03:38:07 | 2026-02-24T13:28:22 | 2026-02-24T13:28:22 | null | CONTRIBUTOR | null | null | {
"total": 0,
"completed": 0,
"percent_completed": 0
} | {
"blocked_by": 0,
"total_blocked_by": 0,
"blocking": 0,
"total_blocking": 0
} | **CI failed on [Unit tests Linux x86-64 pymin_conda_forge_openblas_ubuntu_2204](https://github.com/scikit-learn/scikit-learn/actions/runs/22291890222/job/64480704979)** (Feb 23, 2026)
- test_precomputed_nearest_neighbors_filtering[18] | {
"login": "lesteve",
"id": 1680079,
"node_id": "MDQ6VXNlcjE2ODAwNzk=",
"avatar_url": "https://avatars.githubusercontent.com/u/1680079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lesteve",
"html_url": "https://github.com/lesteve",
"followers_url": "https://api.github.com/users/lesteve/followers",
"following_url": "https://api.github.com/users/lesteve/following{/other_user}",
"gists_url": "https://api.github.com/users/lesteve/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lesteve/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lesteve/subscriptions",
"organizations_url": "https://api.github.com/users/lesteve/orgs",
"repos_url": "https://api.github.com/users/lesteve/repos",
"events_url": "https://api.github.com/users/lesteve/events{/privacy}",
"received_events_url": "https://api.github.com/users/lesteve/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33365/reactions",
"total_count": 1,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33365/timeline | null | completed | null | null | null | false |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33364 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33364/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33364/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33364/events | https://github.com/scikit-learn/scikit-learn/pull/33364 | 3,976,077,453 | PR_kwDOAAzd1s7FiipD | 33,364 | ENH HDBSCAN: add Boruvka MST backend and parallel core-distance queries | {
"login": "gustavorubim",
"id": 7282895,
"node_id": "MDQ6VXNlcjcyODI4OTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/7282895?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gustavorubim",
"html_url": "https://github.com/gustavorubim",
"followers_url": "https://api.github.com/users/gustavorubim/followers",
"following_url": "https://api.github.com/users/gustavorubim/following{/other_user}",
"gists_url": "https://api.github.com/users/gustavorubim/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gustavorubim/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gustavorubim/subscriptions",
"organizations_url": "https://api.github.com/users/gustavorubim/orgs",
"repos_url": "https://api.github.com/users/gustavorubim/repos",
"events_url": "https://api.github.com/users/gustavorubim/events{/privacy}",
"received_events_url": "https://api.github.com/users/gustavorubim/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815435821,
"node_id": "MDU6TGFiZWwxODE1NDM1ODIx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:cluster",
"name": "module:cluster",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 2911057044,
"node_id": "MDU6TGFiZWwyOT... | open | false | [] | null | 7 | 2026-02-23T03:07:46 | 2026-03-06T00:50:00 | null | null | NONE | null | null | null | null | ## Summary
Closes #31503.
This PR adds a Boruvka MST backend to `cluster.HDBSCAN`, exposes configurable MST dispatch via `mst_algorithm`, and parallelizes Boruvka core-distance querying for large datasets.
## Why
On large dense inputs, the HDBSCAN path in scikit-learn lacked a Boruvka MST backend and showed slower fit times than expected compared to the `hdbscan` package.
## What changed
- Added a Boruvka backend implementation:
- `sklearn/cluster/_hdbscan/_boruvka.pyx`
- Added build wiring:
- `sklearn/cluster/_hdbscan/meson.build`
- Added `mst_algorithm={"auto", "brute", "prims", "boruvka_exact", "boruvka_approx"}` and MST dispatch logic:
- `sklearn/cluster/_hdbscan/hdbscan.py`
- Added core-distance query parallelization in Boruvka (`n_jobs` aware split-query path).
- Added non-regression tests and dispatch validation tests:
- `sklearn/cluster/tests/test_hdbscan.py`
- Added ASV coverage for Prim vs Boruvka backends:
- `asv_benchmarks/benchmarks/cluster.py`
- Added changelog entry:
- `doc/whats_new/upcoming_changes/sklearn.cluster/31503.efficiency.rst`
## Performance (local, sklearn-dev env)
Dataset: `make_blobs(n_samples=100000, n_features=20, centers=5, cluster_std=0.6, random_state=0)`
Before this PR:
- `sklearn.HDBSCAN` (pre-change baseline): ~76.4s
After this PR:
- `mst_algorithm="prims"`: ~52.2s
- `mst_algorithm="boruvka_exact"`: ~30.9s
- `mst_algorithm="auto"`: ~33.1s
- `mst_algorithm="boruvka_approx"`: ~4.2s
Comparison to `hdbscan` package on same env:
- `hdbscan` default (approx): ~8.5s
- `hdbscan` exact (`approx_min_span_tree=False`): ~36.2s
Additional profiling:
- Exact-mode runtime is dominated by Boruvka spanning-tree traversal.
- Parallel core-distance query reduced Boruvka init/core-distance time substantially.
## Validation
- `pytest sklearn/cluster/tests/test_hdbscan.py -q`
- 150 passed, 1 warning
- Targeted regression tests for new paths passed.
## Review focus
- Boruvka query split/parallel path in `sklearn/cluster/_hdbscan/_boruvka.pyx`
- API/dispatch semantics in `sklearn/cluster/_hdbscan/hdbscan.py`
- Added tests around algorithm combinations and precomputed constraints
This pull request includes code written with the assistance of AI.
The code has been reviewed by a human.
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33364/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33364/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33364",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33364",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33364.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33364.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33363 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33363/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33363/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33363/events | https://github.com/scikit-learn/scikit-learn/pull/33363 | 3,975,936,263 | PR_kwDOAAzd1s7FiGTl | 33,363 | FIX perplexity normalization for LDA component prior #20032 | {
"login": "gustavorubim",
"id": 7282895,
"node_id": "MDQ6VXNlcjcyODI4OTU=",
"avatar_url": "https://avatars.githubusercontent.com/u/7282895?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/gustavorubim",
"html_url": "https://github.com/gustavorubim",
"followers_url": "https://api.github.com/users/gustavorubim/followers",
"following_url": "https://api.github.com/users/gustavorubim/following{/other_user}",
"gists_url": "https://api.github.com/users/gustavorubim/gists{/gist_id}",
"starred_url": "https://api.github.com/users/gustavorubim/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/gustavorubim/subscriptions",
"organizations_url": "https://api.github.com/users/gustavorubim/orgs",
"repos_url": "https://api.github.com/users/gustavorubim/repos",
"events_url": "https://api.github.com/users/gustavorubim/events{/privacy}",
"received_events_url": "https://api.github.com/users/gustavorubim/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815438551,
"node_id": "MDU6TGFiZWwxODE1NDM4NTUx",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:decomposition",
"name": "module:decomposition",
"color": "151e91",
"default": false,
"description": ""
}
] | open | false | [] | null | 0 | 2026-02-23T01:52:29 | 2026-02-24T17:52:02 | null | null | NONE | null | null | null | null | Fix an inconsistency in LatentDirichletAllocation perplexity where the topic-word prior term was normalized by evaluation corpus size instead of training topic-word mass.
The issue causes nonsensical huge changes in perplexity when evaluating on small splits, because the beta prior term is O(components_) and should be scaled by model size, not evaluation data.
Changes:
- Normalize topic-word prior component in `_perplexity_precomp_distr` using `components_.sum()` when `sub_sampling=False`.
- Add non-regression test checking score/perplexity consistency for this decomposition.
- Add regression test ensuring perplexity behavior is stable between train and small subsets.
Refs #20032.
| null | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33363/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33363/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33363",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33363",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33363.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33363.patch",
"merged_at": null
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33362 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33362/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33362/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33362/events | https://github.com/scikit-learn/scikit-learn/pull/33362 | 3,975,225,141 | PR_kwDOAAzd1s7Ff667 | 33,362 | FIX make it possible to call `dir` on an empty pipeline | {
"login": "jeremiedbb",
"id": 34657725,
"node_id": "MDQ6VXNlcjM0NjU3NzI1",
"avatar_url": "https://avatars.githubusercontent.com/u/34657725?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jeremiedbb",
"html_url": "https://github.com/jeremiedbb",
"followers_url": "https://api.github.com/users/jeremiedbb/followers",
"following_url": "https://api.github.com/users/jeremiedbb/following{/other_user}",
"gists_url": "https://api.github.com/users/jeremiedbb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jeremiedbb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jeremiedbb/subscriptions",
"organizations_url": "https://api.github.com/users/jeremiedbb/orgs",
"repos_url": "https://api.github.com/users/jeremiedbb/repos",
"events_url": "https://api.github.com/users/jeremiedbb/events{/privacy}",
"received_events_url": "https://api.github.com/users/jeremiedbb/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815450942,
"node_id": "MDU6TGFiZWwxODE1NDUwOTQy",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:pipeline",
"name": "module:pipeline",
"color": "151e91",
"default": false,
"description": ""
},
{
"id": 3730476391,
"node_id": "LA_kwDOAAzd1... | closed | false | [] | null | 0 | 2026-02-22T19:01:20 | 2026-02-23T17:01:14 | 2026-02-23T17:01:13 | null | MEMBER | null | null | null | null | Currently, calling `dir` on an empty pipeline fails because some properties try to access the first or last `step` of the pipeline which is not defined for empty pipelines. In such situation, these properties are not defines and should raise an attribute error instead. | {
"login": "ogrisel",
"id": 89061,
"node_id": "MDQ6VXNlcjg5MDYx",
"avatar_url": "https://avatars.githubusercontent.com/u/89061?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ogrisel",
"html_url": "https://github.com/ogrisel",
"followers_url": "https://api.github.com/users/ogrisel/followers",
"following_url": "https://api.github.com/users/ogrisel/following{/other_user}",
"gists_url": "https://api.github.com/users/ogrisel/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ogrisel/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ogrisel/subscriptions",
"organizations_url": "https://api.github.com/users/ogrisel/orgs",
"repos_url": "https://api.github.com/users/ogrisel/repos",
"events_url": "https://api.github.com/users/ogrisel/events{/privacy}",
"received_events_url": "https://api.github.com/users/ogrisel/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33362/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33362/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33362",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33362",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33362.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33362.patch",
"merged_at": "2026-02-23T17:01:13"
} | true |
https://api.github.com/repos/scikit-learn/scikit-learn/issues/33361 | https://api.github.com/repos/scikit-learn/scikit-learn | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33361/labels{/name} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33361/comments | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33361/events | https://github.com/scikit-learn/scikit-learn/pull/33361 | 3,974,817,211 | PR_kwDOAAzd1s7Feo-t | 33,361 | FIX StratifiedGroupKFold raises ValueError when n_splits > n_groups | {
"login": "akuligowski9",
"id": 33849595,
"node_id": "MDQ6VXNlcjMzODQ5NTk1",
"avatar_url": "https://avatars.githubusercontent.com/u/33849595?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/akuligowski9",
"html_url": "https://github.com/akuligowski9",
"followers_url": "https://api.github.com/users/akuligowski9/followers",
"following_url": "https://api.github.com/users/akuligowski9/following{/other_user}",
"gists_url": "https://api.github.com/users/akuligowski9/gists{/gist_id}",
"starred_url": "https://api.github.com/users/akuligowski9/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/akuligowski9/subscriptions",
"organizations_url": "https://api.github.com/users/akuligowski9/orgs",
"repos_url": "https://api.github.com/users/akuligowski9/repos",
"events_url": "https://api.github.com/users/akuligowski9/events{/privacy}",
"received_events_url": "https://api.github.com/users/akuligowski9/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | [
{
"id": 1815446740,
"node_id": "MDU6TGFiZWwxODE1NDQ2NzQw",
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/labels/module:model_selection",
"name": "module:model_selection",
"color": "151e91",
"default": false,
"description": ""
}
] | closed | false | [] | null | 1 | 2026-02-22T15:28:34 | 2026-02-22T23:53:11 | 2026-02-22T23:53:11 | null | NONE | null | null | null | null | ## Summary
Fixes #33085
- `StratifiedGroupKFold` silently produced degenerate (empty) splits when `n_splits` exceeded the number of groups. This adds the same validation check that `GroupKFold` already has, raising a `ValueError` with the same message format.
## Test plan
- [x] Added non-regression test `test_stratified_group_kfold_n_splits_greater_than_n_groups` using the exact reproducer from the issue
- [x] All 45 existing+new StratifiedGroupKFold tests pass
- [x] Pre-commit hooks pass (ruff, mypy, codespell)
## AI disclosure
AI tools (Claude Code) were used to assist with this contribution. All changes have been manually reviewed, tested, and are well understood by the contributor.
🤖 Generated with [Claude Code](https://claude.com/claude-code) | {
"login": "lucyleeow",
"id": 23182829,
"node_id": "MDQ6VXNlcjIzMTgyODI5",
"avatar_url": "https://avatars.githubusercontent.com/u/23182829?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/lucyleeow",
"html_url": "https://github.com/lucyleeow",
"followers_url": "https://api.github.com/users/lucyleeow/followers",
"following_url": "https://api.github.com/users/lucyleeow/following{/other_user}",
"gists_url": "https://api.github.com/users/lucyleeow/gists{/gist_id}",
"starred_url": "https://api.github.com/users/lucyleeow/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/lucyleeow/subscriptions",
"organizations_url": "https://api.github.com/users/lucyleeow/orgs",
"repos_url": "https://api.github.com/users/lucyleeow/repos",
"events_url": "https://api.github.com/users/lucyleeow/events{/privacy}",
"received_events_url": "https://api.github.com/users/lucyleeow/received_events",
"type": "User",
"user_view_type": "public",
"site_admin": false
} | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/issues/33361/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/scikit-learn/scikit-learn/issues/33361/timeline | null | null | null | false | {
"url": "https://api.github.com/repos/scikit-learn/scikit-learn/pulls/33361",
"html_url": "https://github.com/scikit-learn/scikit-learn/pull/33361",
"diff_url": "https://github.com/scikit-learn/scikit-learn/pull/33361.diff",
"patch_url": "https://github.com/scikit-learn/scikit-learn/pull/33361.patch",
"merged_at": null
} | true |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.