Insights: scikit-learn/scikit-learn

1 Release published by 1 person

• 1.2.1 scikit-learn 1.2.1

  published Jan 24, 2023

32 Pull requests merged by 17 people

• TST Allow callables as valid parameter regarding cloning estimator

  #25498 merged Jan 27, 2023

• ENH csr_row_norms optimization

  #24426 merged Jan 27, 2023

• DOC: more didactic examples of bisecting kmeans.

  #25494 merged Jan 27, 2023

• MAINT remove -Wcpp warnings when compiling sklearn.preprocessing._csr_polynomial_expansion

  #25041 merged Jan 27, 2023

• DOC update description of X in `FunctionTransformer.transform()`

  #24844 merged Jan 27, 2023

• CI Add linter display name in Azure

  #25485 merged Jan 26, 2023

• DOC Adds CirrusCI badge to readme

  #25483 merged Jan 26, 2023

• MAINT dynamically expose kulsinski and remove support in BallTree

  #25417 merged Jan 26, 2023

• FIX Support readonly sparse datasets for `manhattan_distances`

  #25432 merged Jan 26, 2023

• MAINT Remove -Wcpp warnings when compiling sklearn.decomposition._online_lda_fast

  #25020 merged Jan 26, 2023

• TST Sets random_state for test_logistic.py

  #25446 merged Jan 25, 2023

• MAINT Update version number to 1.2.1 in SECURITY.md

  #25471 merged Jan 25, 2023

• CI Include linting in CircleCI

  #25475 merged Jan 25, 2023

• MAINT Parameters validation for additive_chi2_kernel

  #25424 merged Jan 24, 2023

• DOC Update news footer for 1.2.1 (#25472)

  #25473 merged Jan 24, 2023

• MAINT Validate parameter for `sklearn.cluster.cluster_optics_xi`

  #25385 merged Jan 24, 2023

• DOC Update news footer for 1.2.1

  #25472 merged Jan 24, 2023

• CI Build doc on CircleCI

  #25466 merged Jan 24, 2023

• TST reduce warnings in test_logistic.py

  #25469 merged Jan 24, 2023

• MAINT refactor spectral_clustering to call SpectralClustering

  #25392 merged Jan 24, 2023

• EFF Improve IsolationForest predict time

  #25186 merged Jan 24, 2023

• ENH Raise NotFittedError in get_feature_names_out for estimators that use ClassNamePrefixFeatureOutMixin and SelectorMixin

  #25308 merged Jan 24, 2023

• Release 1.2.1

  #25442 merged Jan 24, 2023

• FIX faulty test cross_validate that used Lasso while intending to use SVC

  #25456 merged Jan 24, 2023

• DOC Reorder whats_new/v1.2.rst

  #25461 merged Jan 23, 2023

• DOC Make MeanShift documentation clearer

  #25305 merged Jan 23, 2023

• DOC Update date and contributors list for v1.2.1

  #25459 merged Jan 23, 2023

• ENH Raise NotFittedError in get_feature_names_out for MissingIndicator, KBinsDiscretizer, SplineTransformer, DictVectorizer

  #25402 merged Jan 23, 2023

• DOC Fix typo in alpha parameter description

  #25453 merged Jan 23, 2023

• CI Update nogil lock file

  #25455 merged Jan 23, 2023

• FIX propagate configuration to workers in parallel

  #25363 merged Jan 20, 2023

• CI Update nogil lock file

  #25437 merged Jan 20, 2023

15 Pull requests opened by 14 people

• MAINT Refactor Tree Cython class to support modularity

  #25448 opened Jan 20, 2023

• MAINT Improve the `_middle_term_sparse_sparse_{32, 64}` routines

  #25449 opened Jan 21, 2023

• DOC Improve minor and bug-fix release processes' documentation

  #25457 opened Jan 23, 2023

• Added support for ZCA whitening algorithm in _fastica.py

  #25458 opened Jan 23, 2023

• FEA add newton-lsmr solver to LogisticRegression and GLMs

  #25462 opened Jan 23, 2023

• MAINT Parameters validation for sklearn.datasets.fetch_kddcup99

  #25463 opened Jan 23, 2023

• MAINT Parameters validation for sklearn.datasets.make_classification

  #25474 opened Jan 24, 2023

• FIX Enable setting of sub-parameters for deprecated base_estimator param

  #25477 opened Jan 25, 2023

• DOC Add pynndescent to Approximate nearest neighbors in TSNE example

  #25480 opened Jan 25, 2023

• ENH: Update KDTree, and example documentation

  #25482 opened Jan 25, 2023

• ENH allows to overwrite read_csv parameter in fetch_openml

  #25488 opened Jan 26, 2023

• DOC make `max_df` range consistent in CountVectorizer

  #25489 opened Jan 26, 2023

• EFF Speed-up MiniBatchDictionaryLearning by avoiding multiple validation

  #25490 opened Jan 26, 2023

• FIX Issue 25438 Avoiding zeros in NMF β loss calculation causes inconsistency

  #25491 opened Jan 26, 2023

• EFF Speed-up MiniBatchDictionaryLearning by avoiding multiple validation

  #25493 opened Jan 26, 2023

18 Issues closed by 6 people

• OrdinalEncoder with option to mention the start index

  #25400 closed Jan 27, 2023

• Make check_parameters_default_constructible more lenient regarding accepted types

  #25467 closed Jan 27, 2023

• ⚠️ CI failed on Linux_Nightly.pylatest_pip_scipy_dev ⚠️

  #25202 closed Jan 27, 2023

• Partial Dependence Plot orients differently compared to Partial Dependence values

  #25496 closed Jan 27, 2023

• Feature scaling affects decision tree predictions (it shouldn't affect according to the theory)

  #25495 closed Jan 27, 2023

• `FunctionTransformer.transform()` API Confusion

  #24829 closed Jan 27, 2023

• ⚠️ CI failed on Wheel builder ⚠️

  #25497 closed Jan 27, 2023

• test_spectral_embedding_two_components[float32-lobpcg] fails with scipy 1.10

  #25403 closed Jan 26, 2023

• Try to Implemenet the example but error raise (TypeError: __init__() takes 1 positional argument but 3 were given)

  #25468 closed Jan 26, 2023

• circle ci's push_doc.sh fails at git push step

  #25486 closed Jan 26, 2023

• ValueError buffer source array is read-only when computing manahattan_distances of a CSR matrix inside Parallel

  #7981 closed Jan 26, 2023

• Reduce warnings in test_logistic.py

  #24819 closed Jan 24, 2023

• Improving IsolationForest predict time

  #25150 closed Jan 24, 2023

• Make error message uniform when calling `get_feature_names_out` before `fit`

  #24916 closed Jan 24, 2023

• Incorrect test test_validation.py::test_cross_validate?

  #25444 closed Jan 24, 2023

• Error building wheel for scikit-learn 0.21.3

  #25447 closed Jan 21, 2023

• ColumnTransformers don't honor set_config(transform_output="pandas") when multiprocessing with n_jobs>1

  #25239 closed Jan 20, 2023

• ⚠️ CI failed on Linux_nogil.pylatest_pip_nogil ⚠️

  #25430 closed Jan 20, 2023

9 Issues opened by 8 people

• CalibratedClassifierCV doesn't work with `set_config(transform_output="pandas")`

  #25499 opened Jan 27, 2023

• Enable feature selectors to pass pandas DataFrame to estimator

  #25492 opened Jan 26, 2023

• ColumnTransformer with pandas output can't handle transformers with no features

  #25487 opened Jan 26, 2023

• MAINT Replace `cnp.ndarray`s with typed memoryviews

  #25484 opened Jan 26, 2023

• Setting min_samples_split=1 in DecisionTreeClassifier does not raise exception

  #25481 opened Jan 25, 2023

• Regression for wine dataset with pandas parser

  #25478 opened Jan 25, 2023

• AttributeError: 'str' object has no attribute 'set_params' when tuning a boosted classifier

  #25470 opened Jan 24, 2023

• HistGradientBoosting avoid data shuffling when early_stopping activated

  #25460 opened Jan 23, 2023

• Issue building from source on Fedora 37 Python 3.11

  #25452 opened Jan 22, 2023

58 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• ENH add `class_names` in `tree.export_text`

  #25387 commented on Jan 24, 2023 • 18 new comments

• FIX ignore nan values in partial dependence computation

  #25407 commented on Jan 27, 2023 • 10 new comments

• DOC add example to show how to deal with cross-validation

  #18821 commented on Jan 25, 2023 • 9 new comments

• ENH add zero_division parameter to classification metrics

  #23183 commented on Jan 23, 2023 • 6 new comments

• FIX improve convergence criterion for LogisticRegression(penalty="l1", solver='liblinear')

  #25214 commented on Jan 23, 2023 • 6 new comments

• DOC add an example regarding the multiclass strategies

  #25228 commented on Jan 23, 2023 • 6 new comments

• Draft implementation of non-parametric quantile methods (RF, Extra Trees and Nearest Neighbors)

  #19754 commented on Jan 26, 2023 • 5 new comments

• FEA Add Gamma deviance as loss function to HGBT

  #22409 commented on Jan 27, 2023 • 4 new comments

• ENH Adds TargetEncoder

  #25334 commented on Jan 25, 2023 • 4 new comments

• MAINT Remove -Wcpp warnings from sklearn.utils._seq_dataset

  #25406 commented on Jan 27, 2023 • 4 new comments

• MAINT Remove -Wcpp warnings when compiling sklearn.utils.arrayfuncs

  #25415 commented on Jan 24, 2023 • 4 new comments

• RFC Consolidating synchronous communication to one venue

  #25068 commented on Jan 23, 2023 • 3 new comments

• ENH Hellinger distance split criterion for classification trees

  #16478 commented on Jan 27, 2023 • 3 new comments

• FIX BaseLabelPropagation to accept sparse matrices

  #19664 commented on Jan 27, 2023 • 3 new comments

• MAINT Introduce `BaseCriterion` as a base abstraction for `Criterion`s

  #24678 commented on Jan 27, 2023 • 3 new comments

• MAINT Introduce `BaseTree` as a base abstraction for `Tree`

  #25118 commented on Jan 23, 2023 • 3 new comments

• MAINT Parameter validation for sklearn.metrics.d2_pinball_score

  #25414 commented on Jan 24, 2023 • 3 new comments

• DOC Rework Lasso and Elastic Net for Sparse Signals example

  #25429 commented on Jan 23, 2023 • 3 new comments

• FIX MLPEstimator respect max_iter when warm_start is on

  #25443 commented on Jan 25, 2023 • 3 new comments

• Avoiding zeros in NMF β loss calculation causes inconsistency

  #25438 commented on Jan 26, 2023 • 2 new comments

• Trees with MAE criterion are slow to train

  #9626 commented on Jan 27, 2023 • 2 new comments

• Add Warnings when trying to use estimator.fit() functions with Pandas objects with different indices.

  #25419 commented on Jan 27, 2023 • 2 new comments

• ENH Adds missing value support for trees

  #23595 commented on Jan 24, 2023 • 2 new comments

• ENH Allow for appropriate dtype us in `preprocessing.PolynomialFeatures` for sparse matrices

  #23731 commented on Jan 27, 2023 • 2 new comments

• MAINT Parameter Validation for `covariance.graphical_lasso`

  #25053 commented on Jan 24, 2023 • 2 new comments

• MAINT Refactor `Splitter` into a `BaseSplitter`

  #25101 commented on Jan 26, 2023 • 2 new comments

• NDCG score doesn't work with binary relevance and a list of 1 element

  #21335 commented on Jan 21, 2023 • 1 new comment

• PLSRegression VIP score calculation

  #7050 commented on Jan 22, 2023 • 1 new comment

• MLPEstimator do not report the proper `n_iter_` after successive call of `fit`

  #24764 commented on Jan 24, 2023 • 1 new comment

• Math not rendered in downloaded documentation when offline

  #20951 commented on Jan 25, 2023 • 1 new comment

• Metrics for prediction intervals

  #20162 commented on Jan 25, 2023 • 1 new comment

• ENH partial_dependece plot for HistGradientBoosting estimator fitted with `sample_weight`

  #25210 commented on Jan 26, 2023 • 1 new comment

• MAINT Use Cython memoryviews

  #10624 commented on Jan 26, 2023 • 1 new comment

• Support typing

  #16705 commented on Jan 26, 2023 • 1 new comment

• LogisticRegression code is way to opaque

  #11865 commented on Jan 26, 2023 • 1 new comment

• FEA post-fit calibration option in HGBT

  #22435 commented on Jan 27, 2023 • 1 new comment

• Make automatic validation for all scikit-learn public functions

  #24862 commented on Jan 27, 2023 • 1 new comment

• Open up `check_array` and `BaseEstimator._validate_data` to overriding `xp.asarray` with an additional callable parameter `asarray_fn`

  #25433 commented on Jan 27, 2023 • 1 new comment

• Added vip_ attribute to PLSRegression

  #13492 commented on Jan 23, 2023 • 1 new comment

• [MRG] ENH Adds categories='dtypes' option to OrdinalEncoder and OneHotEncoder

  #15396 commented on Jan 26, 2023 • 1 new comment

• MRG Logistic regression preconditioning

  #15583 commented on Jan 24, 2023 • 1 new comment

• API Deprecates values in partial_dependence in favor of pdp_values

  #21809 commented on Jan 27, 2023 • 1 new comment

• [MRG] PERF Significant performance improvements in Partial Least Squares (PLS) Regression

  #23876 commented on Jan 21, 2023 • 1 new comment

• [DRAFT] Engine plugin API and engine entry point for Lloyd's KMeans

  #24497 commented on Jan 27, 2023 • 1 new comment

• ENH Make GMM initialization more robust and easily user customizeable

  #24812 commented on Jan 21, 2023 • 1 new comment

• ENH Automatic handling of categorical columns in Hist Gradient Boosting models

  #24907 commented on Jan 26, 2023 • 1 new comment

• MAINT Parameters validation for spectral_clustering

  #25378 commented on Jan 24, 2023 • 1 new comment

• MAINT Parameters validation for reconstruct_from_patches_2d

  #25384 commented on Jan 24, 2023 • 1 new comment

• Feature Request: function to calculate Expected Calibration Error (ECE)

  #18268 commented on Jan 23, 2023 • 0 new comments

• MAINT Remove all Cython, C and C++ compilations warnings

  #24875 commented on Jan 26, 2023 • 0 new comments

• There is no raise expectation

  #25421 commented on Jan 26, 2023 • 0 new comments

• ⚠️ CI failed on Linux_Nightly_PyPy.pypy3 ⚠️

  #25117 commented on Jan 27, 2023 • 0 new comments

• FEA Confusion matrix derived metrics

  #19556 commented on Jan 26, 2023 • 0 new comments

• ENH add Naive Bayes Metaestimator `ColumnwiseNB` (aka "GeneralNB")

  #22574 commented on Jan 26, 2023 • 0 new comments

• Add sample_weight to the calculation of alphas in enet_path and LinearModelCV

  #23045 commented on Jan 26, 2023 • 0 new comments

• PERF Implement `PairwiseDistancesReduction` backend for `KNeighbors.predict_proba`

  #24076 commented on Jan 27, 2023 • 0 new comments

• feat: Support class weight and sample weight for MLPClassifier

  #25326 commented on Jan 26, 2023 • 0 new comments

• [WIP] ENH Support sample weights when fitting HistGradientBoosting estimator

  #25431 commented on Jan 26, 2023 • 0 new comments