39 Pull requests merged by 19 people

• ENH check_array returns numeric array w/ mixed typed dataframes

  #22237 merged Apr 24, 2022

• MNT Adds black commit to git-blame-ignore-revs

  #23208 merged Apr 24, 2022

• FIX Add missing f prefix for f-string in check_sample_weights_not_overwritten

  #23206 merged Apr 24, 2022

• MNT Combine multiple `isinstance` call

  #23204 merged Apr 24, 2022

• DOC Small fixes in Tweedie deviances documentation

  #22952 merged Apr 23, 2022

• DOC Improve symbol convention in `model_evaluation.rst`

  #23201 merged Apr 23, 2022

• DOC Consistently present BinaryTree method signatures without self

  #17734 merged Apr 23, 2022

• DOC Fixes grammar in decision tree's cpp pruning docstring

  #20442 merged Apr 23, 2022

• DOC Adds missing B to the denominator in model_evaluation

  #23202 merged Apr 23, 2022

• CLN Remove unnecessary code from DecisionBoundaryDisplay

  #23192 merged Apr 23, 2022

• DOC Fixes docstring in BayesianGaussianMixture

  #19243 merged Apr 23, 2022

• DOC Fixes confusing notation for y and y_pred

  #23193 merged Apr 23, 2022

• CI Fix don't run all random seeds all the time

  #23190 merged Apr 22, 2022

• [MRG] Online implementation of non-negative matrix factorization

  #16948 merged Apr 22, 2022

• DOC Ensure that normalize passes numpydoc validation

  #23188 merged Apr 22, 2022

• TST Speed-up test_minibatch_dictionary_learning_dtype_match

  #23186 merged Apr 22, 2022

• DOC use notebook-style for plot_rbm_logistic_classification.py

  #23104 merged Apr 22, 2022

• TST make test_ridge_regression_hstacked_X more lenient

  #23170 merged Apr 22, 2022

• DOC Fix notebook style of plot_coin_ward_segmentation

  #23164 merged Apr 21, 2022

• TST Fix ridge tests not deterministic

  #23181 merged Apr 21, 2022

• DOC Rework plot_ard.py example

  #22747 merged Apr 21, 2022

• ENH improve the linear models inspection example

  #23082 merged Apr 20, 2022

• CI Do not auto close tracking issue if tests pass

  #23155 merged Apr 20, 2022

• TST Fix failing tests catched by azure cron job

  #23167 merged Apr 20, 2022

• DOC fix: update docstring to say onehotencoder handles missing values

  #23054 merged Apr 20, 2022

• nogil: fix race condition when running k-means without the GIL

  #23159 merged Apr 20, 2022

• make test_ridge_regression_vstacked_X less sensitive

  #23152 merged Apr 20, 2022

• TST adapt tol for ridge tests to pass on all random seeds

  #23017 merged Apr 20, 2022

• CI allow to run only selected tests but on all random seeds

  #23026 merged Apr 20, 2022

• DOC Ensure `dump_svmlight_file` passes numpydoc validation

  #23166 merged Apr 20, 2022

• MNT Replace pytest.warns(None) in test_neighbors

  #23142 merged Apr 19, 2022

• FIX: fetch covtype dataset concurrent-safe

  #23113 merged Apr 19, 2022

• CI enable global_dtype fixture on nightly build jobs

  #23110 merged Apr 19, 2022

• MNT Replace pytest.warns(None) in linear_models tests

  #23138 merged Apr 19, 2022

• MNT Replace pytest.warns(None) in test_utils

  #23137 merged Apr 19, 2022

• MNT Replace pytest.warns(None) in test_text.py

  #23136 merged Apr 19, 2022

• FIX Fix available_if with bounded methods

  #23077 merged Apr 19, 2022

• FIX Fixes Stacking* HTML repr

  #23097 merged Apr 19, 2022

• Fix rst title in support documentation

  #23157 merged Apr 19, 2022

13 Pull requests opened by 9 people

• [FEAT] Implement quantile SVR

  #23153 opened Apr 18, 2022

• DOC Improve the documentation of log loss for multi-label cases

  #23154 opened Apr 18, 2022

• Fixed multiple tar slips from using "extractall" which can lead to an exploit

  #23163 opened Apr 19, 2022

• CI Experimental [nogil] build of scikit-learn

  #23174 opened Apr 21, 2022

• [FEAT] Zero division nan

  #23183 opened Apr 21, 2022

• DOC use notebook-style in ensemble/plot_adaboost_hastie_10_2.py

  #23184 opened Apr 22, 2022

• DOC Adds svg logo without subtitle

  #23191 opened Apr 22, 2022

• ENH Imporves error message when experimental flag is not imported

  #23194 opened Apr 22, 2022

• WIP ENH Cythonize `_assert_all_finite` using stop-on-first strategy

  #23197 opened Apr 22, 2022

• DOC Maintain a list of all estimators that supports missing data

  #23198 opened Apr 23, 2022

• ENH Using context manager for file operation

  #23207 opened Apr 24, 2022

• DOC use notebook-style in linear_model/plot_ridge_path.py

  #23209 opened Apr 24, 2022

• ENH propagate `eigen_tol` to all eigen solver

  #23210 opened Apr 24, 2022

41 Issues closed by 7 people

• Bug in train_test_split

  #22979 closed Apr 24, 2022

• Missing `f` prefix on f-strings

  #23203 closed Apr 24, 2022

• Add post-prune option for decision tree as well as randomForest and GBDT?

  #4630 closed Apr 24, 2022

• Y-hat is used unconventionally within documentation for model evaluation

  #20538 closed Apr 23, 2022

• KMeans.transform should allow a metric optional parameter

  #3214 closed Apr 23, 2022

• Bayesian Gaussian Mixture Models

  #19023 closed Apr 23, 2022

• User guide formulas for macro average recall and precision seem wrong

  #18864 closed Apr 23, 2022

• Retrieve feature importances for any model or any that SelectFromModel supports

  #9110 closed Apr 23, 2022

• Implement inverse_transform method for cross_decomposition ?

  #6439 closed Apr 23, 2022

• Incorrect error bar predicted by scikit-learn

  #14395 closed Apr 22, 2022

• [regression] --record {plist} records a directory

  #12681 closed Apr 22, 2022

• SVD decomposition float32 float64 stability

  #13309 closed Apr 22, 2022

• Interpolation of ROC and PR curve metrics

  #18135 closed Apr 22, 2022

• average_precision_score() overestimates AUC value

  #13074 closed Apr 22, 2022

• RandomForestClassifier parallel issues with CPU usage decreasing over run

  #12482 closed Apr 22, 2022

• IndexError: index is out of bounds

  #12037 closed Apr 22, 2022

• Index structure creation for knn-classifier...

  #11826 closed Apr 22, 2022

• Logistic regression producing inconsistent results when warm starting with `solver='sag'` and `solver='saga'`.

  #11511 closed Apr 22, 2022

• check_supervised_y_2d fails with estimator GaussianProcess on travis with macOS machines

  #11439 closed Apr 22, 2022

• Low cpu utilization when running OneVsRestClassifier with n_jobs=-1

  #11448 closed Apr 22, 2022

• set_params fn for pipeline to maintain order of parameters

  #10439 closed Apr 22, 2022

• use extras_require in setup.py

  #11361 closed Apr 22, 2022

• The azure CI times out more and more often

  #23185 closed Apr 22, 2022

• Online implementation of Non-negative Matrix Factorizarion (NMF)

  #13308 closed Apr 22, 2022

• Different results depending on labels using linear_svc and adaboost

  #23187 closed Apr 22, 2022

• test_ridge_regression_hstacked_X is not stable

  #23169 closed Apr 22, 2022

• test_ridge_regression_unpenalized is unstable

  #23177 closed Apr 22, 2022

• Pipeline: apply all transformations except the last classifier

  #8414 closed Apr 22, 2022

• Import Error when importing from different threads

  #13940 closed Apr 22, 2022

• For SVM with precomputed kernel, ROC AUC scores are inconsistent, when using cross_val_score and predict_proba

  #6266 closed Apr 22, 2022

• CalibratedClassifierCV allowing for more types of input data

  #23176 closed Apr 22, 2022

• ndcg_score

  #23175 closed Apr 21, 2022

• website repo close to its limit

  #21171 closed Apr 21, 2022

• ⚠️ CI failed on macOS.pylatest_conda_mkl_no_openmp ⚠️

  #23165 closed Apr 20, 2022

• Doc Examples Bug: `plot_column_transformer_mixed_types`, missing categorical SimpleImputer

  #23053 closed Apr 20, 2022

• The score evaluation with "roc_auc" is not the same in GridSearchCV() and roc_auc_score() functon

  #23158 closed Apr 20, 2022

• test_ridge_regression_vstacked_X is not stable

  #23014 closed Apr 20, 2022

• Covtype dataset raises error when fetching

  #23048 closed Apr 19, 2022

• Getting ETA for model training

  #23156 closed Apr 19, 2022

• Cannot pickle predict_proba function with 1.0 release

  #21344 closed Apr 19, 2022

• ⚠️ CI failed on Linux_Runs.pylatest_conda_forge_mkl ⚠️

  #23151 closed Apr 19, 2022

9 Issues opened by 7 people

• Exposing submodules directly from main sklearn module instead of recursively importing each submodule separately

  #23205 opened Apr 24, 2022

• Consider LSMR instead of LSQR

  #23199 opened Apr 23, 2022

• Add more robust Kmeans initializer for mixture.GaussianMixture

  #23195 opened Apr 22, 2022

• Investigate SAG/SAGA solver

  #23180 opened Apr 21, 2022

• Wrong normalization for pseudolikelihood in Restricted Boltzmann Machine.

  #23179 opened Apr 21, 2022

• roc_auc_score must warn user about using labels instead of probabilities in binary case

  #23178 opened Apr 21, 2022

• Clearify and test dropping categories in linear models

  #23172 opened Apr 20, 2022

• MinCovDet estimation of covariance with strong bias?

  #23162 opened Apr 19, 2022

• Compare BIRCH and MiniBatchKMeans - unused variable

  #23160 opened Apr 19, 2022

81 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.

• FEA Bisecting K-Means

  #20031 commented on Apr 22, 2022 • 24 new comments

• DOC Adds release highlights for 1.1

  #23084 commented on Apr 23, 2022 • 14 new comments

• CI: move Linux and MacOS Azure builds to conda lock files

  #22448 commented on Apr 25, 2022 • 12 new comments

• DEP calibration_curve normalize parameter in calibration.py

  #23095 commented on Apr 23, 2022 • 12 new comments

• [WIP] ENH: Hellinger distance split criterion for classification trees

  #16478 commented on Apr 24, 2022 • 10 new comments

• DOC: use notebook-style for plot_covariance_estimation.py

  #23150 commented on Apr 20, 2022 • 7 new comments

• ENH improve the random forest feature importance comparison example

  #23085 commented on Apr 21, 2022 • 6 new comments

• DOC Add example comparing permutation importance and SHAP

  #18139 commented on Apr 22, 2022 • 5 new comments

• FEA add PredictionErrorDisplay

  #18020 commented on Apr 21, 2022 • 3 new comments

• FET add retry mechanism for fetch_xx functions

  #21691 commented on Apr 23, 2022 • 3 new comments

• FIX `cross_validate` with multimetric scoring returns the non-failed scorers results if there is a failing scorer

  #23101 commented on Apr 24, 2022 • 3 new comments

• ENH Optimize runtime for IsolationForest

  #23149 commented on Apr 24, 2022 • 3 new comments

• Add DateTimeTransformer?

  #21968 commented on Apr 19, 2022 • 2 new comments

• SLEP006 - equal requests? {fit, partial_fit}, {predict, predict_proba, predict_log_proba, decision_function}

  #22988 commented on Apr 20, 2022 • 2 new comments

• Link to logos in Community section of website

  #23072 commented on Apr 23, 2022 • 2 new comments

• ENH Adds Array API support to LinearDiscriminantAnalysis

  #22554 commented on Apr 21, 2022 • 2 new comments

• Are predictions on scikit-learn models thread-safe?

  #18530 commented on Apr 18, 2022 • 1 new comment

• `_check_feature_names` raises UserWarning when accesing bagged estimators

  #21599 commented on Apr 19, 2022 • 1 new comment

• Add rolling window to sklearn.model_selection.TimeSeriesSplit

  #22523 commented on Apr 19, 2022 • 1 new comment

• Add zero_division None or np.nan

  #22625 commented on Apr 19, 2022 • 1 new comment

• Add PAV algorithm for calibration_curve/reliability diagrams

  #23132 commented on Apr 20, 2022 • 1 new comment

• BIC and AIC scores for the Bayesian Gaussian Mixture class

  #19523 commented on Apr 20, 2022 • 1 new comment

• KDTree's query & query_radius methods have inconsistent return values

  #5976 commented on Apr 22, 2022 • 1 new comment

• TSNE with correlation metric: ValueError: Distance matrix 'X' must be symmetric

  #4475 commented on Apr 22, 2022 • 1 new comment

• DOC maintain a list of all estimators that natively handle missing data

  #21382 commented on Apr 23, 2022 • 1 new comment

• Ensure that functions's docstrings pass numpydoc validation

  #21350 commented on Apr 24, 2022 • 1 new comment

• Fix notebook-style examples

  #22406 commented on Apr 24, 2022 • 1 new comment

• [WIP] Add Absolute Mean Percentage Error as available loss function in SGDR…

  #6605 commented on Apr 24, 2022 • 1 new comment

• ENH propagate `eigen_tol` to all eigen solver

  #11968 commented on Apr 19, 2022 • 1 new comment

• Run itrees in parallel during prediction.

  #14001 commented on Apr 24, 2022 • 1 new comment

• MNT small refactor _bagging.py

  #15673 commented on Apr 22, 2022 • 1 new comment

• [WIP] BUG returned most regularized model with Ridge*CV in case of tie

  #16029 commented on Apr 24, 2022 • 1 new comment

• [MRG] DOC Adds 'scoring' to plot_learning_curve example

  #16900 commented on Apr 23, 2022 • 1 new comment

• ENH allow to pass str or scorer to make_scorer

  #18141 commented on Apr 24, 2022 • 1 new comment

• [MRG]Adding an argument to make it easy to change font size of values in plot_confusion_matrix (Fixes #19051)

  #19070 commented on Apr 23, 2022 • 1 new comment

• MNT Adds CurveDisplayMixin _get_response_values

  #20999 commented on Apr 19, 2022 • 1 new comment

• MNT refactor _get_response_values

  #21538 commented on Apr 19, 2022 • 1 new comment

• MAINT rename base_estimator to estimator in CalibratedClassifierCV

  #22054 commented on Apr 24, 2022 • 1 new comment

• ENH Add eigh as a solver in MDS

  #22330 commented on Apr 22, 2022 • 1 new comment

• FEA add gamma loss to HGBT

  #22409 commented on Apr 20, 2022 • 1 new comment

• ENH Feature: calculate normed stress (Stress-1) in `manifold.MDS`

  #22562 commented on Apr 22, 2022 • 1 new comment

• ENH Added `dtype` preservation to `Birch`

  #22968 commented on Apr 22, 2022 • 1 new comment

• ENH use more blas functions in cd solvers

  #22972 commented on Apr 19, 2022 • 1 new comment

• Using callable in RandomForest oob_score parameter to define the oob scoring strategy.

  #23043 commented on Apr 25, 2022 • 1 new comment

• FET add argument `n_targets` for GPR

  #23099 commented on Apr 22, 2022 • 1 new comment

• Scikit-learn office hours Monday May 2, 2022 from 4 pm to 5 pm (UTC) on the scikit-learn discord channel

  #23070 commented on Apr 19, 2022 • 0 new comments

• PolynomialFeatures doesn't work correctly when degree=0

  #19551 commented on Apr 21, 2022 • 0 new comments

• Inconsistent density estimator sampling APIs

  #10539 commented on Apr 22, 2022 • 0 new comments

• sparse-sparse and sparse-dense dot product with dense output

  #516 commented on Apr 23, 2022 • 0 new comments

• Create Examples / Tests for custom estimators not inheriting from BaseEstimator

  #1265 commented on Apr 23, 2022 • 0 new comments

• Cross validation and tools for undersampling and oversampling

  #1362 commented on Apr 23, 2022 • 0 new comments

• Relevance Vector Machine (RVM)

  #1513 commented on Apr 23, 2022 • 0 new comments

• Numerical stability of symmetric decorrelation in FastICA

  #2735 commented on Apr 23, 2022 • 0 new comments

• Calibration and Refinement loss for Brier score loss

  #21774 commented on Apr 24, 2022 • 0 new comments

• Deprecate normalize parameter in `calibration_curve`

  #22482 commented on Apr 25, 2022 • 0 new comments

• [MRG] Matern kernel

  #3885 commented on Apr 23, 2022 • 0 new comments

• ENH Adds typing support to LogisticRegression

  #17799 commented on Apr 24, 2022 • 0 new comments

• Update _mds.py

  #18094 commented on Apr 22, 2022 • 0 new comments

• ENH Extend PDP for nominal categorical features

  #18298 commented on Apr 22, 2022 • 0 new comments

• FIX Ridge.coef_ return a single dimension array when target type is not continuous-multiple

  #19746 commented on Apr 22, 2022 • 0 new comments

• FIX VotingClassifier handles class weights

  #19753 commented on Apr 23, 2022 • 0 new comments

• ENH Adds class_weight to HistGradientBoostingClassifier

  #22014 commented on Apr 24, 2022 • 0 new comments

• [MRG] Add a separator between a criteria and other contents in a node in export_graphviz for visibility

  #22022 commented on Apr 24, 2022 • 0 new comments

• ENH added original version of pickled estimator in state dict

  #22094 commented on Apr 24, 2022 • 0 new comments

• MAINT rename base_estimator to estimator in BaggingClassifier and BaggingRegressor

  #22145 commented on Apr 22, 2022 • 0 new comments

• DOC Improve the mathematical description of Logistic Regression

  #22382 commented on Apr 22, 2022 • 0 new comments

• FEA Add positive and negative likelihood ratios to metrics

  #22518 commented on Apr 25, 2022 • 0 new comments

• ENH Add `eigh` solver to `FastICA`

  #22527 commented on Apr 22, 2022 • 0 new comments

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

  #22574 commented on Apr 23, 2022 • 0 new comments

• Clarify group order in GroupKFold and LeaveOneGroupOut

  #22582 commented on Apr 22, 2022 • 0 new comments

• Include `HDBSCAN` as a sub-module for `sklearn.cluster`

  #22616 commented on Apr 23, 2022 • 0 new comments

• MAINT float 32bit support for `DistanceMetric`

  #22764 commented on Apr 20, 2022 • 0 new comments

• DOC Use notebook style in plot_kernel_ridge_regression.py

  #22804 commented on Apr 20, 2022 • 0 new comments

• Deprecate `class_weight_` and `n_support_` in regression/single class models in SVM module

  #22898 commented on Apr 22, 2022 • 0 new comments

• DOC rework plot_document_classification_20newsgroups.py example

  #22928 commented on Apr 25, 2022 • 0 new comments

• [MRG] Add sparse input support to OPTICS

  #22965 commented on Apr 22, 2022 • 0 new comments

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

  #23045 commented on Apr 21, 2022 • 0 new comments

• MAINT Use check_scalar to validate scalar in: BayesianRidge

  #23051 commented on Apr 25, 2022 • 0 new comments

• ENH Cythonize `dump_svmlight_file`

  #23127 commented on Apr 22, 2022 • 0 new comments

• DOC: Fixing notebook-style formatting on plot_segmentation_toy.py

  #23140 commented on Apr 20, 2022 • 0 new comments

• Update plot_pca_vs_fa_model_selection.py

  #23148 commented on Apr 20, 2022 • 0 new comments