scikit-learn / scikit-learn

27 Pull requests merged by 21 people

• DOC Fixed typo in cross_val_predict docstring

  #19739 merged Mar 21, 2021

• DOC Fix load iris datasets

  #19729 merged Mar 21, 2021

• DOC: Fix typo in KDE metric docstring default value

  #19735 merged Mar 21, 2021

• DOC Fix doc for single linkage in feature agglomeration

  #19715 merged Mar 21, 2021

• DOC Update broken link in conftest.py

  #19736 merged Mar 21, 2021

• Remove assert warn from preprocessing tests

  #19691 merged Mar 20, 2021

• Stratified Group KFold implementation

  #18649 merged Mar 20, 2021

• DOC Fix doc regarding required_parameters

  #19725 merged Mar 19, 2021

• DOC Move Sacred to "Experimentation frameworks"

  #19730 merged Mar 19, 2021

• DOC Add scikit-survival to related projects

  #19728 merged Mar 19, 2021

• Replaced assert_raises from utils/tests/test_estimator_checks

  #19709 merged Mar 19, 2021

• Removed assert_warns_message from gaussian_process/tests

  #19697 merged Mar 19, 2021

• [MRG] ENH Consistent loss name for squared error

  #19310 merged Mar 19, 2021

• DOC Correct scorer documentation

  #19720 merged Mar 19, 2021

• Fix typo in elastic net docstring.

  #19711 merged Mar 18, 2021

• MNT move PolynomialFeatures from _data.py to _polynomial.py

  #19611 merged Mar 18, 2021

• ENH Better error for corrupted files in fetch_kddcup99

  #19669 merged Mar 18, 2021

• Added utils.gen_batches in documentation, Solving issue #19681

  #19688 merged Mar 17, 2021

• MAINT Python 3.9 in badge on README.rst

  #19702 merged Mar 17, 2021

• CI Use conda instead of pip to install anaconda-client

  #19528 merged Mar 17, 2021

• ENH Add periodic extrapolation to SplineTransformer

  #19483 merged Mar 17, 2021

• [MRG] Fix documentation for russelrao formula

  #19695 merged Mar 17, 2021

• Add a test to check the consistency of the Ridge and ElasticNet(l1_ratio=0) solutions

  #19620 merged Mar 17, 2021

• FIX RuntimeWarning division by zero in check_classifiers_one_label

  #19690 merged Mar 16, 2021

• TST Change assert from sklearn to pytest style in tests/test_pipeline.py

  #19678 merged Mar 16, 2021

• ENH Adds _num_features for array-likes

  #19633 merged Mar 16, 2021

• FIX RuntimeWarning by dividing by zero in test_iforest_with_uniform_data

  #19622 merged Mar 15, 2021

18 Pull requests opened by 17 people

• Pass the confusion matrix as a parameter

  #19686 opened Mar 16, 2021

• DOC Improve documentation consistency for GaussianProcessRegressor

  #19687 opened Mar 16, 2021

• ENH Adds n_features_in_ checking to multioutput

  #19692 opened Mar 16, 2021

• Prevent division by zero in GPR when y_train is constant

  #19703 opened Mar 17, 2021

• MAINT Improve issue template

  #19704 opened Mar 17, 2021

• [FIX] GaussianProcessRegression(normalize_y=True).predict(X, return_cov=True)

  #19706 opened Mar 17, 2021

• ENH BestSplitter split on feature with lowest index when multiple features have identical order

  #19707 opened Mar 18, 2021

• check_random_state moved up in order to use it in make_dataset and improve reproduction

  #19716 opened Mar 18, 2021

• DOC unify None format of the metrics

  #19718 opened Mar 18, 2021

• skip check_classifiers_predictions test for poor_scoor estimators

  #19719 opened Mar 19, 2021

• [MRG] Fix top_k_accuracy_score ignoring labels for "multiclass" case

  #19721 opened Mar 19, 2021

• [MRG] Fix encoder categories having dtype='S'

  #19727 opened Mar 19, 2021

• ENH Improve regularization messages for QuadraticDiscriminantAnalysis

  #19731 opened Mar 19, 2021

• FIX a bug in KernelPCA.inverse_transform

  #19732 opened Mar 20, 2021

• ENH Consistent loss name for absolute error

  #19733 opened Mar 20, 2021

• [MRG] more efficient _num_combinations calculation

  #19734 opened Mar 20, 2021

• ENH num_features for a 1d collection of dicts is undefined

  #19740 opened Mar 21, 2021

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

  #19746 opened Mar 22, 2021

16 Issues closed by 11 people

• FeatureAgglomeration documentation error

  #15259 closed Mar 21, 2021

• Stratified GroupKFold

  #13621 closed Mar 20, 2021

• Mean absolute scaled error for forecasting evaluation

  #18685 closed Mar 19, 2021

• 0.20.4 sources on PyPI include generated .c files from 0.21.x or master

  #15006 closed Mar 19, 2021

• CircleCI keeps failing on master

  #13287 closed Mar 19, 2021

• MAINT Move PolynomialFeatures to _polynomial.py

  #19574 closed Mar 18, 2021

• Just initialize a neural network without fitting

  #19701 closed Mar 18, 2021

• Corrupted Loaded Dataset Returns Unhelpful Error

  #19667 closed Mar 18, 2021

• utils.gen_batches missing from documentation

  #19681 closed Mar 17, 2021

• try to downgrade the sklearn version and rise a strange error

  #19689 closed Mar 17, 2021

• [BUG?] OverflowError in _liblinear.npx

  #19660 closed Mar 17, 2021

• PCA Score function returning positive values

  #19696 closed Mar 17, 2021

• Documentation for russellrao appears incorrect

  #19694 closed Mar 17, 2021

• "RuntimeWarning: invalid value encountered in true_divide" in tests

  #19334 closed Mar 16, 2021

• '<' not supported between instances of 'tuple' and 'Version' error

  #19672 closed Mar 15, 2021

• ValueError in _lars_path_solver in _least_angle.py

  #19682 closed Mar 15, 2021

9 Issues opened by 9 people

• Make Isomap to work with any precomputed neighborhood graph, at least radius_neighbors_graph

  #19744 opened Mar 22, 2021

• [Changed behavior] n_iter_no_change should be attached with early_stopping, not model

  #19743 opened Mar 22, 2021

• BUG: Regression with StandardScaler due to #19527

  #19726 opened Mar 19, 2021

• Transform method for TSNE (different from the fit_transform already present)

  #19717 opened Mar 18, 2021

• Introduce honest sampling into sklearn RandomForestClassifier

  #19710 opened Mar 18, 2021

• [RFC] Minimal scipy version for 1.0 (or 0.26) release

  #19705 opened Mar 17, 2021

• [RFC] Removing "Other" template in favour of Github Discussions?

  #19700 opened Mar 17, 2021

• Adding min-max linkage to Agglomerative Clustering

  #19698 opened Mar 17, 2021

• Ridge and Lasso return different shaped .coef_ attributes

  #19693 opened Mar 16, 2021

66 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.

• [MRG] Add quantile regression

  #9978 commented on Mar 22, 2021 • 45 new comments

• Test and doc for n_features_in_ for sklearn.calibration

  #19555 commented on Mar 21, 2021 • 17 new comments

• [MRG] Online linear One-Class SVM using SGD implementation

  #10027 commented on Mar 21, 2021 • 16 new comments

• [MRG after #12145] Add "Randomized SVD" solver option to KernelPCA for faster partial decompositions, like in PCA

  #12069 commented on Mar 18, 2021 • 10 new comments

• [WIP] FIX index overflow error in sparse matrix polynomial expansion (bis)

  #19676 commented on Mar 17, 2021 • 6 new comments

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

  #16948 commented on Mar 18, 2021 • 5 new comments

• Improve initialization and learning rate in t-SNE

  #19491 commented on Mar 21, 2021 • 5 new comments

• FIX Fixes regression in CCA

  #19646 commented on Mar 19, 2021 • 5 new comments

• display='diagram' needs more prominence in documentation

  #18305 commented on Mar 18, 2021 • 4 new comments

• SplineTransformer raises when passing knots explicity for n_knots=None

  #19598 commented on Mar 21, 2021 • 4 new comments

• Handling GradientBoostingClassifier Over-shooting

  #11452 commented on Mar 21, 2021 • 4 new comments

• [MRG] fixes #14257 - Group aware Time-based cross validation

  #16236 commented on Mar 18, 2021 • 4 new comments

• [MRG] ENH Add `trust-ncg` option to LogisticRegression

  #17877 commented on Mar 19, 2021 • 4 new comments

• Prevent division by zero in GPR when y_train is constant

  #18388 commented on Mar 17, 2021 • 4 new comments

• Confusion matrix derived metrics

  #19556 commented on Mar 18, 2021 • 4 new comments

• Adaboost Learning Rate Documentation

  #19521 commented on Mar 19, 2021 • 3 new comments

• [WIP] sample props (proposal 4)

  #16079 commented on Mar 17, 2021 • 3 new comments

• Handling division by zero std for GaussianProcessRegressor

  #19361 commented on Mar 19, 2021 • 3 new comments

• Give feedback when `svm.SVC` is configured with kernel hyperparameters for a different kernel

  #19614 commented on Mar 15, 2021 • 2 new comments

• Pass the confusion matrix as a parameter

  #19679 commented on Mar 16, 2021 • 2 new comments

• Remove the use of assert_warns and assert_warns_message from the tests

  #19414 commented on Mar 16, 2021 • 2 new comments

• roc_auc_score min_tpr parameter?

  #11668 commented on Mar 17, 2021 • 2 new comments

• GaussianProcessRegressor doesn't work with multidemensional output when normalize_y=True

  #18065 commented on Mar 18, 2021 • 2 new comments

• decision_path method for GradientBoosting

  #19294 commented on Mar 20, 2021 • 2 new comments

• Overflow error following standard scalar in LinearSVC: value to large to convert to npy_int32

  #18027 commented on Mar 20, 2021 • 2 new comments

• OneHot/OrdinalEncoder categories broken for dtype='S'

  #19677 commented on Mar 21, 2021 • 2 new comments

• [WIP] Model trees (M5P)

  #13732 commented on Mar 20, 2021 • 2 new comments

• [MRG] FEA Gower distance

  #16834 commented on Mar 21, 2021 • 2 new comments

• DOC Added new scientific reference of MCC in the _classification.py page

  #18707 commented on Mar 17, 2021 • 2 new comments

• [WIP] DOC Display diagram to pipeline example

  #18758 commented on Mar 18, 2021 • 2 new comments

• Add memory=joblib.Memory param to OPTICS

  #19024 commented on Mar 17, 2021 • 2 new comments

• EXA improve example of forest feature importances digits

  #19429 commented on Mar 19, 2021 • 2 new comments

• ENH Add Poisson criterion to RandomForestRegressor

  #19464 commented on Mar 16, 2021 • 2 new comments

• Warn when gamma is set with linear kernel (#19614)

  #19630 commented on Mar 15, 2021 • 2 new comments

• [MRG] PLS target values (Y) inverse-transform

  #19680 commented on Mar 21, 2021 • 2 new comments

• Variable regularisation strengths for Lasso Ridge and ElasticNet

  #19656 commented on Mar 15, 2021 • 1 new comment

• sklearn.datasets.load_files select file extension

  #19269 commented on Mar 15, 2021 • 1 new comment

• Order of features selected in Sequential feature selection transformer

  #19583 commented on Mar 15, 2021 • 1 new comment

• Quantile Regression Forest [Feature request]

  #11086 commented on Mar 16, 2021 • 1 new comment

• GaussianProcessRegressor cannot correctly predict std in multi-target scene

  #17394 commented on Mar 17, 2021 • 1 new comment

• How to get eigenvectors of kernel pca

  #17171 commented on Mar 17, 2021 • 1 new comment

• Should cross-validation scoring take sample-weights into account?

  #4632 commented on Mar 18, 2021 • 1 new comment

• Add "other" / min_frequency option to OneHotEncoder

  #12153 commented on Mar 19, 2021 • 1 new comment

• LogisticRegression memory consumption goes crazy on 0.22+

  #17125 commented on Mar 19, 2021 • 1 new comment

• Clearify group order in GroupKFold and LeaveOneGroupOut

  #18338 commented on Mar 19, 2021 • 1 new comment

• SimpleImputer, missing_values and None

  #19071 commented on Mar 20, 2021 • 1 new comment

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

  #18530 commented on Mar 21, 2021 • 1 new comment

• Unexpected behavior when passing multiple parameter sets to RandomizedSearchCV

  #18057 commented on Mar 21, 2021 • 1 new comment

• [WIP] FEA New meta-estimator to post-tune the decision_function/predict_proba threshold for binary classifiers

  #16525 commented on Mar 19, 2021 • 1 new comment

• Callbacks API

  #16925 commented on Mar 16, 2021 • 1 new comment

• [MRG] LinearRegression Optimizations

  #17560 commented on Mar 16, 2021 • 1 new comment

• FIX reshape the predicitons array shape in IterativeImputer

  #19409 commented on Mar 17, 2021 • 1 new comment

• ENH Checks n_features_in_ after fitting in mixture

  #19540 commented on Mar 21, 2021 • 1 new comment

• GaussianMixture with BIC/AIC

  #19562 commented on Mar 17, 2021 • 1 new comment

• Differences in penalties handled by solvers in Logistic Regression Documentation

  #19651 commented on Mar 15, 2021 • 0 new comments

• Regression in GP standard deviation where y_train.std() == 0

  #18318 commented on Mar 17, 2021 • 0 new comments

• Make HistGradientBoostingRegressor/Classifer use _openmp_effective_n_threads to set the default maximum number of threads to use

  #16016 commented on Mar 19, 2021 • 0 new comments

• HistGradientBoosting* interaction constraints

  #19148 commented on Mar 19, 2021 • 0 new comments

• Bias correction for TransformedTargetRegressor

  #15881 commented on Mar 19, 2021 • 0 new comments

• Track SLEP10: Add n_features_in_ to all modules

  #19333 commented on Mar 21, 2021 • 0 new comments

• QDA Classifier: divide by zero if data not of full rank

  #14997 commented on Mar 22, 2021 • 0 new comments

• [WIP] Common test for equivalence between sparse and dense matrices.

  #13246 commented on Mar 17, 2021 • 0 new comments

• [MRG] Improve R2 and Explained Variance scores by making it clearer that by default there is a fix for constant true y

  #17266 commented on Mar 20, 2021 • 0 new comments

• DOC Add example comparing permutation importance and SHAP

  #18139 commented on Mar 22, 2021 • 0 new comments

• Record output of transformers in ColumnTransformer

  #18393 commented on Mar 17, 2021 • 0 new comments

• MNT Applies black formatting to most of the code base

  #18948 commented on Mar 19, 2021 • 0 new comments