scikit-learn / scikit-learn Public

48 Pull requests merged by 29 people

• DOC fix shape of output in _BaseNB._joint_log_likelihood docstring

  #22487 merged Feb 15, 2022

• MNT Update black to stable version

  #22474 merged Feb 15, 2022

• FIX skip tests when estimator has non_deterministic tag

  #22318 merged Feb 15, 2022

• DOC Ensures that fetch_openml passes numpydoc validation

  #22483 merged Feb 15, 2022

• BUG Fix covariance and stdev shape in GPR with normalize_y

  #22199 merged Feb 14, 2022

• ENH preserves dtypes for GenericUnivariateSelect class

  #22370 merged Feb 14, 2022

• REV Revert check finite in discriminant_analysis

  #22457 merged Feb 14, 2022

• DOC use notebook-style for example plot_feature_selection.py

  #22437 merged Feb 14, 2022

• ENH Loss module LogisticRegression

  #21808 merged Feb 14, 2022

• DOC Fix notebook-style for plot_document_clustering.py

  #22443 merged Feb 14, 2022

• DOC use notebook-style for plot_train_error_vs_test_error

  #22440 merged Feb 14, 2022

• [MGR] DOC Ensures that fetch_olivetti_faces passes numpydoc validation

  #22480 merged Feb 14, 2022

• DOC fix section marker in plot_ard.py examples

  #22481 merged Feb 14, 2022

• DOC Fix MDS images in user guide

  #22465 merged Feb 14, 2022

• ENH optimize memory usage for `datasets.make_s_curve`

  #22479 merged Feb 14, 2022

• DOC Ensures that assert_all_finite passes numpydoc validation

  #22470 merged Feb 14, 2022

• API Lowercase estimator argument in check_estimator()

  #22188 merged Feb 14, 2022

• DOC: use notebook-style in plot_prediction_latency.py

  #22418 merged Feb 14, 2022

• DOC Ensures that isotonic_regression passes numpydoc validation

  #22475 merged Feb 14, 2022

• DOC Ensures that radius_neighbors_graph passes numpydoc

  #22462 merged Feb 12, 2022

• DOC Ensure that kneighbors_graph passes numpydoc validation

  #22459 merged Feb 12, 2022

• ENH avoid checking for finite in ldsr and svd call in LinearDiscriminantAnalysis

  #18909 merged Feb 12, 2022

• DOC Fix duplicate-word typos

  #22454 merged Feb 12, 2022

• DOC Ensures that function passes numpydoc validation: paired_distances

  #22380 merged Feb 11, 2022

• MAINT Adjust tests for numpydoc 1.2

  #22287 merged Feb 11, 2022

• ENH Better error message for check_scalar

  #22218 merged Feb 11, 2022

• CI remove macOS compilers constraint

  #22451 merged Feb 11, 2022

• MAINT Copy latest version of file from upstream

  #22447 merged Feb 11, 2022

• DOC Correct docstring for RocCurveDisplay.from_predictions

  #22444 merged Feb 10, 2022

• MNT Run dos2unix

  #21053 merged Feb 10, 2022

• MAINT Plug `PairwiseDistancesArgKmin` as a back-end

  #22288 merged Feb 10, 2022

• ENH reduce memory usage in `make_blobs`

  #22412 merged Feb 10, 2022

• FIX Preserve float32 in MiniBatchDictionaryLearning

  #22428 merged Feb 10, 2022

• DOC use notebook-style for plot_lasso_and_elasticnet

  #22423 merged Feb 10, 2022

• DOC Clarify splitting behavior of `fit_params`

  #22379 merged Feb 9, 2022

• DOC Ensures that load_digits passes numpydoc validation

  #22392 merged Feb 9, 2022

• DOC Fixes plot_sgd example to be consistent with 0.24

  #22424 merged Feb 9, 2022

• MAINT: remove SKLEARN_SITE_JOBLIB doc which has been a noop since 0.21

  #22429 merged Feb 9, 2022

• MNT remove _pairwise attribute

  #21594 merged Feb 9, 2022

• DOC Ensures that make_scorer passes numpydoc validation

  #22367 merged Feb 9, 2022

• Update plot_regression.py

  #22416 merged Feb 9, 2022

• MNT Bump joblib version dependency to 1.0.0

  #22365 merged Feb 9, 2022

• TST Complete test for `pairwise_distance_{argmin,argmin_min}`

  #22371 merged Feb 8, 2022

• MNT remove deprecated attributes in PLS

  #21585 merged Feb 8, 2022

• DOC fix SciPy intersphinx link.

  #22419 merged Feb 8, 2022

• TST Improved testing logic in `tests/test_logistic.py`

  #22314 merged Feb 8, 2022

• DOC change adrin's sponsor to Hugging Face

  #22415 merged Feb 8, 2022

• FIX Only raise feature name warning with mixed types and strings

  #22410 merged Feb 8, 2022

15 Pull requests opened by 15 people

• Kernel Ridge Classifier

  #22417 opened Feb 8, 2022

• Correct precision at k

  #22431 opened Feb 10, 2022

• Store pairs of classes used to train each estimator in OneVsOneClassifier

  #22436 opened Feb 10, 2022

• [WIP] CI Enable cython coverage

  #22439 opened Feb 10, 2022

• First attempt at using lock files in the CI

  #22448 opened Feb 11, 2022

• DOC use notebook-style for plot_faces_decomposition

  #22452 opened Feb 11, 2022

• DOC Update _graph_lasso.py

  #22456 opened Feb 12, 2022

• DOC FIX notebook-style for plot_stock_market

  #22461 opened Feb 12, 2022

• DOC Ensures that load_wine passes numpydoc

  #22469 opened Feb 13, 2022

• DOC Ensures that precision_recall_fscore_support passes numpydoc validation

  #22472 opened Feb 13, 2022

• API Deprecate max_feature=`auto` for tree classes

  #22476 opened Feb 14, 2022

• DOC Ensures that sklearn.datasets._base.load_linnerud passes numpydoc validation

  #22484 opened Feb 14, 2022

• [WIP] separate penalty factors for GLM regressions

  #22485 opened Feb 15, 2022

• FIX Fix DummyRegressor overriding constant

  #22486 opened Feb 15, 2022

• DOC: put more emphasis on r2_score

  #22491 opened Feb 15, 2022

14 Issues closed by 5 people

• pairwise_distances fails with metric='seuclidean' if input is not of type np.double

  #15731 closed Feb 15, 2022

• Multi-target GPR predicts only 1 std when normalize_y=False

  #22174 closed Feb 14, 2022

• Multi-target GPR sample_y fails when normalize_y=True

  #22175 closed Feb 14, 2022

• spectral_embedding tests fail on 64-bit Little Endian PowerPC (ppc64le)

  #16794 closed Feb 14, 2022

• Broken image in MDS user guide documentation

  #22061 closed Feb 14, 2022

• Clarification around `cross_val_predict` cross validator.

  #22477 closed Feb 14, 2022

• Return correlation in f_regression function

  #19566 closed Feb 13, 2022

• add which solver was used in sklearn Ridge regression : add clf.get_params(solver)

  #22455 closed Feb 12, 2022

• Adding a White Kernel to GP Regressor makes predictions all 0

  #22445 closed Feb 11, 2022

• convergence criteria in semi-supervised learning

  #11877 closed Feb 10, 2022

• Performance regression in MiniBatchDictionaryLearning

  #22426 closed Feb 10, 2022

• ⚠️ CI failed on Linux_Nightly.pylatest_pip_scipy_dev ⚠️

  #22364 closed Feb 9, 2022

• Identity Mapping

  #22422 closed Feb 9, 2022

• Do not deprecate tuple support for feature names: [FutureWarning: Feature names only support names that are all strings Got feature names with dtypes: ['tuple']. An error will be raised in 1.2.]

  #22399 closed Feb 8, 2022

22 Issues opened by 18 people

• main branch has failing test case for tree module

  #22490 opened Feb 15, 2022

• Should not use `collections.abc.Sequence` to ducktype `ndarray`

  #22489 opened Feb 15, 2022

• OneHotEncoder creates non unique column names, when values None (NoneType) and "None" (str) are both present

  #22488 opened Feb 15, 2022

• CalibrationDisplay.from_predictions() does not have a 'normalize' argument

  #22482 opened Feb 14, 2022

• DummyRegressor converts some params to NumPy after fit()

  #22478 opened Feb 14, 2022

• Model Persistence page is missing side navigation contents

  #22473 opened Feb 13, 2022

• The weighted average should be replaced with a weighted sum?

  #22466 opened Feb 12, 2022

• Workflow Improvement/Clarification: Issue vs. PR

  #22463 opened Feb 12, 2022

• Deprecate max_features="auto" in DecisionTree*

  #22458 opened Feb 12, 2022

• Sensitivity Analysis function

  #22453 opened Feb 11, 2022

• `test_y_multioutput`in Gaussian Process is failing on Debian 32bit

  #22446 opened Feb 11, 2022

• StandardScaler and PolynomialFeatures fail on zero-feature inputs during fit (should become passthrough)

  #22442 opened Feb 10, 2022

• Verbosity for OneVsRestClassifier

  #22441 opened Feb 10, 2022

• Path for pluggable low-level computational routines

  #22438 opened Feb 10, 2022

• FEA post-fit calibration option in HGBT

  #22435 opened Feb 10, 2022

• PCA hangs occasionally when applied to data that fits in memory

  #22434 opened Feb 10, 2022

• DOC correct use of SAMPLE

  #22433 opened Feb 10, 2022

• Some ideas for a developer-level API

  #22432 opened Feb 10, 2022

• GPR `sample_y` enforce `n_targets=1` before calling `fit`

  #22430 opened Feb 9, 2022

• Improve the metrics API

  #22427 opened Feb 9, 2022

• Pin dependencies version in CI with lock files with a bot that auto-updates the lockfiles

  #22425 opened Feb 9, 2022

• Y_trans is not equal to y_scores_ in PLSRegression

  #22420 opened Feb 8, 2022

124 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.

• Base sample-prop implementation and docs (alternative to #21284)

  #22083 commented on Feb 15, 2022 • 19 new comments

• ENH Allow prefit in stacking

  #22215 commented on Feb 15, 2022 • 17 new comments

• Solve Gaussian Mixture Model sample/kmeans++-based initialisation merge conflicts #11101

  #20408 commented on Feb 15, 2022 • 15 new comments

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

  #16948 commented on Feb 10, 2022 • 13 new comments

• [WIP] DOC Add minimal reproducer guide for scikit-learn

  #22401 commented on Feb 9, 2022 • 13 new comments

• Path for Adopting the Array API spec

  #22352 commented on Feb 14, 2022 • 12 new comments

• ENH Add get_feature_names_out for RandomTreesEmbedding module

  #21762 commented on Feb 15, 2022 • 12 new comments

• Add data_transposed argument and warning to make_sparse_coded_signal

  #21425 commented on Feb 14, 2022 • 11 new comments

• DOC Improve the mathematical description of Logistic Regression

  #22382 commented on Feb 13, 2022 • 11 new comments

• TST invalid init parameters for losses

  #22407 commented on Feb 14, 2022 • 11 new comments

• FEA add gamma loss to HGBT

  #22409 commented on Feb 9, 2022 • 10 new comments

• Fix notebook-style examples

  #22406 commented on Feb 14, 2022 • 9 new comments

• Ensure that functions's docstrings pass numpydoc validation

  #21350 commented on Feb 14, 2022 • 9 new comments

• ENH Support CSR matrix in type_of_target

  #14862 commented on Feb 12, 2022 • 9 new comments

• MRG stacking classifier support multilabel classification.

  #20424 commented on Feb 10, 2022 • 9 new comments

• Add attribute with target class pairs of each estimator in OneVsOneClassifier

  #22366 commented on Feb 14, 2022 • 8 new comments

• ENH improve ARFF parser using pandas

  #21938 commented on Feb 11, 2022 • 7 new comments

• ValueError when a key present in class_weight is not found in labels

  #22413 commented on Feb 15, 2022 • 6 new comments

• Randomized PCA.transform uses a lot of RAM

  #11102 commented on Feb 15, 2022 • 6 new comments

• Isomap supports radius-based neighbors

  #19794 commented on Feb 15, 2022 • 5 new comments

• FIX FunctionTransformer check_input fails for object type input

  #19916 commented on Feb 15, 2022 • 4 new comments

• [MRG] ENH speed up plot_poisson_regression_non_normal_loss.py

  #21787 commented on Feb 14, 2022 • 4 new comments

• add which solver was used in sklearn Ridge regression : add clf.get_params(solver)

  #22341 commented on Feb 9, 2022 • 3 new comments

• Deprecate the parameter `**kwds` in function `def paired_distances` since it is unused

  #22408 commented on Feb 10, 2022 • 3 new comments

• Add friedman's H statistic

  #22383 commented on Feb 13, 2022 • 3 new comments

• Draft implementation of quantile methods

  #19754 commented on Feb 10, 2022 • 3 new comments

• ENH Introduce `PairwiseDistancesReduction` and `PairwiseDistancesArgKmin` (feature branch)

  #22134 commented on Feb 10, 2022 • 3 new comments

• DOC Ensures that function passes numpydoc validation: f1_score

  #22358 commented on Feb 9, 2022 • 3 new comments

• GraphicalLasso incorrect description

  #22325 commented on Feb 10, 2022 • 2 new comments

• Pooled variance for LDA estimators

  #22411 commented on Feb 10, 2022 • 2 new comments

• Improving the runtime efficiency of the SVC for a precomputed kernel during inference

  #22363 commented on Feb 10, 2022 • 2 new comments

• Net benefit curve and decision curve analysis

  #22136 commented on Feb 10, 2022 • 2 new comments

• Default hyperparameter values for novelty detection with LocalOutlierFactor and OneClassSVM

  #22402 commented on Feb 11, 2022 • 2 new comments

• Enable IP Anonymization for the Google Analytics tracker of scikit-learn.org

  #11499 commented on Feb 14, 2022 • 2 new comments

• MNT small refactor _bagging.py

  #15673 commented on Feb 12, 2022 • 2 new comments

• [MRG] Added option to return raw predictions from cross_validate

  #18390 commented on Feb 11, 2022 • 2 new comments

• ENH kind accept a list of str in plot_partial_dependence

  #19438 commented on Feb 13, 2022 • 2 new comments

• ENH Adds file extension selection to load_files

  #19747 commented on Feb 15, 2022 • 2 new comments

• Add custom_range argument for partial dependence

  #21033 commented on Feb 15, 2022 • 2 new comments

• ENH Pairwise Distances ArgKmin

  #21462 commented on Feb 11, 2022 • 2 new comments

• FET add retry mechanism for fetch_xx functions

  #21691 commented on Feb 15, 2022 • 2 new comments

• DOC Clarify the LS term in example

  #22156 commented on Feb 15, 2022 • 2 new comments

• ENH Optimise `decomposition.FastICA.fit` memory footprint and runtime

  #22268 commented on Feb 14, 2022 • 2 new comments

• ENH Allow `SelectFromModel`'s `max_features` to accept callables

  #22356 commented on Feb 13, 2022 • 2 new comments

• sklearn MDS vs skbio PCoA

  #15272 commented on Feb 8, 2022 • 1 new comment

• sklearn.utils.resample stratify parameter is not working

  #17321 commented on Feb 9, 2022 • 1 new comment

• RFC bump up dependencies (numpy, scipy and python) minimum versions for scikit-learn 1.1

  #21460 commented on Feb 9, 2022 • 1 new comment

• Add Positive Likelihood Ratio (and negative) to metrics

  #22404 commented on Feb 10, 2022 • 1 new comment

• Add positive / nonnegative constraint to ARDRegression

  #22172 commented on Feb 10, 2022 • 1 new comment

• Change name of Triage Team

  #21993 commented on Feb 11, 2022 • 1 new comment

• The default value of some parameters make little sense

  #13570 commented on Feb 11, 2022 • 1 new comment

• Implement GeneralNB

  #15077 commented on Feb 11, 2022 • 1 new comment

• .fit.transform != .fit_transform inconsistency in PCA results

  #18941 commented on Feb 13, 2022 • 1 new comment

• red/green scatter plots. Colorblindness friendlyness in Documentation

  #5435 commented on Feb 14, 2022 • 1 new comment

• Use the function check_scalar for parameters validation

  #21927 commented on Feb 14, 2022 • 1 new comment

• Use check_scalar for parameters validation

  #20724 commented on Feb 15, 2022 • 1 new comment

• [MRG] Added colorblind compatibility for plot examples in calibration

  #5538 commented on Feb 14, 2022 • 1 new comment

• Added colorblind compatibility cluster examples

  #5555 commented on Feb 14, 2022 • 1 new comment

• Warning the user of bad default values, starting by dbscan.eps

  #14942 commented on Feb 11, 2022 • 1 new comment

• [WIP] Implement general naive Bayes

  #16281 commented on Feb 11, 2022 • 1 new comment

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

  #16478 commented on Feb 14, 2022 • 1 new comment

• [MRG] Refactor MiniBatchDictionaryLearning and add stopping criterion

  #18975 commented on Feb 11, 2022 • 1 new comment

• MNT move an orphaned vendored file around

  #21166 commented on Feb 14, 2022 • 1 new comment

• Bugfix: Fixed Attribut self.n_components reference

  #21533 commented on Feb 9, 2022 • 1 new comment

• ENH Add inverse_transform to random projection transformers

  #21701 commented on Feb 8, 2022 • 1 new comment

• [ENH] Introduce Manifold-regularized Discriminative NMF

  #21763 commented on Feb 14, 2022 • 1 new comment

• FEA add quantile HGBT

  #21800 commented on Feb 11, 2022 • 1 new comment

• DOC Maintain a list of all estimators that natively handle missing data #21382

  #22096 commented on Feb 14, 2022 • 1 new comment

• ENH Preserving dtype for np.float32 in LatentDirichletAllocation

  #22113 commented on Feb 14, 2022 • 1 new comment

• [MRG] Adding variable `force_alpha` to classes in naive_bayes.py

  #22269 commented on Feb 13, 2022 • 1 new comment

• ENH Implement SVD algorithm in MDS

  #22330 commented on Feb 14, 2022 • 1 new comment

• Preserving dtype for float32 / float64 in transformers

  #11000 commented on Feb 9, 2022 • 0 new comments

• Documentation is silent about non-metric MDS 0's being interpreted as missing data

  #22391 commented on Feb 10, 2022 • 0 new comments

• Add Blog to top menu bar

  #22337 commented on Feb 10, 2022 • 0 new comments

• Major Bug with SpectralClustering

  #22160 commented on Feb 10, 2022 • 0 new comments

• power_t: does it make sense for this parameter to have negative values

  #22178 commented on Feb 10, 2022 • 0 new comments

• Incorrect implementation of homogeneity, completeness and v-measure

  #22185 commented on Feb 10, 2022 • 0 new comments

• Sphinx docs don't build correctly on Windows 10

  #22232 commented on Feb 10, 2022 • 0 new comments

• Calibration and Refinement loss for Brier score loss

  #21774 commented on Feb 10, 2022 • 0 new comments

• DecisionTreeClassifier behaviour when there are 2 or more best splitter (a tie among splitters)

  #12259 commented on Feb 14, 2022 • 0 new comments

• Text vectorizers clean-up / TfidfVectorizer deprecation

  #14951 commented on Feb 14, 2022 • 0 new comments

• Implement `get_feature_names_out` for all estimators

  #21308 commented on Feb 15, 2022 • 0 new comments

• sklearn.datasets.load_files select file extension

  #19269 commented on Feb 15, 2022 • 0 new comments

• Fix the issue #3745, the code book generation for OutputCodeClassifier

  #3768 commented on Feb 8, 2022 • 0 new comments

• dict_vectorizer: added inverse_onehot to original feature names + updatable vectorizer dict

  #3776 commented on Feb 9, 2022 • 0 new comments

• Implemented the H-Measure metric for binary classification.

  #4034 commented on Feb 8, 2022 • 0 new comments

• [MRG] GBM & meta-ensembles - support for class_weight

  #4215 commented on Feb 14, 2022 • 0 new comments

• Add monotonicity parameter to Gradient Boosting and Decision Trees

  #4950 commented on Feb 14, 2022 • 0 new comments

• [WIP] Default dict for GridSearch.

  #5564 commented on Feb 9, 2022 • 0 new comments

• [MRG] Support multi-threading of LibLinear L1 one-vs-rest LogisticRegression for # classes > 2

  #6448 commented on Feb 14, 2022 • 0 new comments

• [MRG] LabelPropagation and LabelSpreading enhancements

  #6604 commented on Feb 14, 2022 • 0 new comments

• [WIP] Progress Logger

  #8105 commented on Feb 14, 2022 • 0 new comments

• [WIP] Allow SGDClassifier to support np.float32 without upcasting to float64

  #9084 commented on Feb 14, 2022 • 0 new comments

• ENH: new method to return all intermediate activations in MLP

  #9469 commented on Feb 14, 2022 • 0 new comments

• different activation functions at each hidden layer of MLP

  #10097 commented on Feb 14, 2022 • 0 new comments

• Fixes #4632: Allow cross validation to have weighted test scores

  #10806 commented on Feb 14, 2022 • 0 new comments

• ENH: add oob support to CalibratedClassifier

  #11175 commented on Feb 14, 2022 • 0 new comments

• [WIP] API: allow cross validation to work with partial_fit

  #11266 commented on Feb 14, 2022 • 0 new comments

• Add Multiple Correspondance Analysis, PCA for categorical data

  #11627 commented on Feb 14, 2022 • 0 new comments

• ENH: Add support for cosine distance in k-means

  #12192 commented on Feb 14, 2022 • 0 new comments

• [WIP] priority of features in decision to splitting

  #12364 commented on Feb 14, 2022 • 0 new comments

• [MRG] Support multi-label probability calibration

  #13060 commented on Feb 14, 2022 • 0 new comments

• [Feature] Normalized Precision-Recall Metrics Added

  #13113 commented on Feb 14, 2022 • 0 new comments

• ENH: Preserve float32/64 for SGD

  #13346 commented on Feb 14, 2022 • 0 new comments

• Weighted scoring in cross validation (Closes #4632)

  #13432 commented on Feb 14, 2022 • 0 new comments

• Tree export fill colors

  #13490 commented on Feb 14, 2022 • 0 new comments

• Warn if eps is not specificed for DBSCAN

  #13541 commented on Feb 11, 2022 • 0 new comments

• Fuzzy c-means clustering algorithm is partly implemented.

  #13669 commented on Feb 14, 2022 • 0 new comments

• [WIP] - Precision and recall at k

  #14859 commented on Feb 14, 2022 • 0 new comments

• Deprecate TfidfVectorizer

  #14966 commented on Feb 14, 2022 • 0 new comments

• [MRG] Fix LDA with only one class

  #17403 commented on Feb 9, 2022 • 0 new comments

• DOC Add example comparing permutation importance and SHAP

  #18139 commented on Feb 10, 2022 • 0 new comments

• [MRG] added Code of Conduct

  #18989 commented on Feb 9, 2022 • 0 new comments

• DOC Added references to the Matthews correlation coefficient function in the user guide

  #19958 commented on Feb 14, 2022 • 0 new comments

• New Feature - Bisecting K-Means

  #20031 commented on Feb 14, 2022 • 0 new comments

• DOC Improve documentation regarding some pitfalls in interpretation

  #20451 commented on Feb 14, 2022 • 0 new comments

• MNT Adds quicker CI failure to reduce resource usage

  #21497 commented on Feb 12, 2022 • 0 new comments

• FIX make the shape of LogisticRegression(solver="liblinear") consistent with other solvers

  #21998 commented on Feb 15, 2022 • 0 new comments

• [MRG] Add D2 pinbal score and D2 absolute error score

  #22118 commented on Feb 13, 2022 • 0 new comments

• DOC Update documentation of gradient boosting estimators w/ ranges

  #22153 commented on Feb 14, 2022 • 0 new comments

• FIX Supports numpy strings and array-like in KMean.init

  #22154 commented on Feb 11, 2022 • 0 new comments

• [MRG] [ENH] Adds a `return_std_of_f` flag for GPCs

  #22227 commented on Feb 11, 2022 • 0 new comments

• [MRG] Update gridsearch example for multimetric scoring.

  #22279 commented on Feb 13, 2022 • 0 new comments

• MNT Add validation for parameter alphas in `LassoCV`

  #22305 commented on Feb 15, 2022 • 0 new comments