scikit-learn/scikit-learn

24 Pull requests merged by 14 people

• Merged #14003 [MRG] CLN Only one function for parallel predict Jun 3, 2019
• Merged #13997 MAINT: test_encoder_dtypes_pandas reads expected dtype from DF Jun 3, 2019
• Merged #13991 [MRG] Doctest with print change only adjusts default options for doctest Jun 1, 2019
• Merged #13987 [MRG] Avoid uncessary copies in sklearn.preprocessing Jun 1, 2019
• Merged #13999 DOC Fix typo: omit comma May 31, 2019
• Merged #13993 Fixed documentation typos May 31, 2019
• Merged #13946 [MRG][DOC] Fix inconsistencies in clustering doc. May 30, 2019
• Merged #13919 CLN GBDTs: don't split on last bin (explicitly) May 30, 2019
• Merged #13899 adds an example to OneClassSVM May 30, 2019
• Merged #13983 Added matplotlib to show_versions() May 29, 2019
• Merged #13980 FIX Lazy cython import for pytest to work without cython May 29, 2019
• Merged #13839 [MRG] DEP change the default of cv and n_splits May 29, 2019
• Merged #13975 FIX use CD solver in face decomposition to use postive parameter May 29, 2019
• Merged #13855 [MRG] DEP remove legacy mode from OneHotEncoder May 29, 2019
• Merged #13838 [MRG] DEP change default and deprecate normalize_components in SparsePCA May 29, 2019
• Merged #13863 [WIP] DEP remove lars positive parameter May 29, 2019
• Merged #13970 MAINT: use explicit value of n_jobs to avoid hangs on Windows May 28, 2019
• Merged #13966 [MRG] CLN Refactors code May 28, 2019
• Merged #13927 CLN Removed max_bins from splitter in GBDT May 28, 2019
• Merged #13963 PERF Free problem and param sooner in liblinear.train_wrap May 28, 2019
• Merged #13704 [MRG] Changed VarianceThreshold behaviour when threshold is zero. See #13691 May 28, 2019
• Merged #13933 FIX Bin training and validation data separately in GBDTs May 28, 2019
• Merged #13955 [MRG] Remove parent negative loss calculation from for loop to improve performance May 28, 2019
• Merged #13959 [MRG] DOC: fix contributing guide section ordering May 28, 2019

11 Pull requests proposed by 9 people

• Proposed #13960 [MRG] Allow sparse input to incremental PCA May 27, 2019
• Proposed #13961 [MRG] Contributing guide update May 28, 2019
• Proposed #13974 [MRG] Initialize MissingIndicator with error_on_new = False May 29, 2019
• Proposed #13978 MAINT: Per #13977 set non-default absolute tolerance May 29, 2019
• Proposed #13981 Add blurb to landing page May 29, 2019
• Proposed #13982 set body max width for sphinx template May 29, 2019
• Proposed #13988 Avoid memory copies in linear_model when fit_intercept=False May 30, 2019
• Proposed #13995 switch to sparse_cg solver in Ridge when X is sparse and fit_intercept is True May 31, 2019
• Proposed #14001 Run itrees in parallel during prediction. May 31, 2019
• Proposed #14004 MAINT Use isinstance(x, numbers.Integral) to check for integer dtype May 31, 2019
• Proposed #14005 [MRG] Fix 'SparseSeries deprecated: scipy-dev failing on travis' #14002 May 31, 2019

16 Issues closed by 6 people

• Closed #13992 sklearn/preprocessing/tests/test_encoders.py::test_encoder_dtypes_pandas fails on Windows Jun 3, 2019
• Closed #13508 sklearn/utils/tests/test_pprint.py is too brittle to API changes Jun 1, 2019
• Closed #13564 can put NORMALIZE_WHITESPACE in pytest config Jun 1, 2019
• Closed #13985 should doctests use print_changed_only=True? Jun 1, 2019
• Closed #14006 'No module named sklearn.impute' Jun 1, 2019
• Closed #6205 solve underflow errors by changing utils/extmath.py -> logsumexp() May 31, 2019
• Closed #8920 multiprocessing.pool.MaybeEncodingError: Error sending result May 31, 2019
• Closed #9440 Test error in test_feature_importances with current master on fedora 26. May 31, 2019
• Closed #5709 msvc compiler - building on win7-64bit, python-2.7-32bit May 31, 2019
• Closed #7851 Suggested Installation Docs for Ubuntu May 31, 2019
• Closed #13973 Add matplotlib to show_versions() May 29, 2019
• Closed #4649 [RFC] Use explicit module levels skip lists for estimator checks May 28, 2019
• Closed #13691 VarianceThreshold doesn't remove feature with zero variance May 28, 2019
• Closed #13926 GBDTs should bin train and validation data separately? May 28, 2019
• Closed #13931 In histogram GBDTs, the parent node negative loss should be computed only once May 28, 2019
• Closed #13965 Why not support Mahalanobis distance for k-means ? May 28, 2019

23 Issues created by 17 people

• Opened #14011 Add Matrix Profile algorithm to scikit-learn Jun 2, 2019
• Opened #14010 Strange output from sklearn.manifold.Isomap Jun 2, 2019
• Opened #14008 RFC Set print_changed_only=True by default for meta-estimators Jun 1, 2019
• Opened #14007 'cannot import name '_object_dtype_isnan' Jun 1, 2019
• Opened #14002 SparseSeries deprecated: scipy-dev failing on travis May 31, 2019
• Opened #14000 Isolation Forest executes single-threaded during prediction May 31, 2019
• Opened #13998 `RidgeCV` `cv_values_` are for preprocessed data: centered and scaled by sample weights. May 31, 2019
• Opened #13994 Kernel Linear Discriminant Analysis May 30, 2019
• Opened #13990 add three-fold split method train/val/test May 30, 2019
• Opened #13989 GaussianProcessRegressor multi-output underlying mechanics? May 30, 2019
• Opened #13986 Avoid copy=True by default in check_array May 29, 2019
• Opened #13984 Ignore `setup.py` files in codecov May 29, 2019
• Opened #13979 part of the landing page is in layout.html May 29, 2019
• Opened #13977 Default atol value in assert_allclose_dense_sparse is too low May 29, 2019
• Opened #13976 IndexError: list index out of range in export_text when the tree only has one feature May 29, 2019
• Opened #13972 how can cv folds can be more than number of groups in cross_validate? May 29, 2019
• Opened #13971 no rotation and no arrows drew betwee nodes in plot_tree May 29, 2019
• Opened #13969 Add non-strict mode to check_estimator May 28, 2019
• Opened #13968 Design of add_indicator in SimpleImputer may fail when running cross validation May 28, 2019
• Opened #13967 Feature request: warm starting for histogram-based GBM May 28, 2019
• Opened #13964 Experimental error message May 28, 2019
• Opened #13962 Use of target_names in example datasets is inconsistent May 28, 2019
• Opened #13957 IncrementalPCA should accept sparse input May 27, 2019

50 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.

• 59 new comments Open #13146 [MRG] Adds Permutation Importance May 31, 2019
• 29 new comments Open #13911 [MRG] Native support for missing values in GBDTs Jun 2, 2019
• 23 new comments Open #13954 [MRG]Allow Gaussian process kernels on structured data #13936 May 31, 2019
• 22 new comments Open #13938 [MRG] Added fit and score times for learning curves May 30, 2019
• 21 new comments Open #12852 [MRG] Adds KNNImputer May 30, 2019
• 20 new comments Open #13806 [MRG] Add function score_samples to Pipeline (fix issue #12542) May 31, 2019
• 19 new comments Open #12789 [MRG] Adds multiclass ROC AUC May 31, 2019
• 14 new comments Open #12569 [MRG] Add Categorical Naive Bayes Jun 2, 2019
• 13 new comments Open #13923 Linear regression unnecessarily slow May 30, 2019
• 11 new comments Open #13891 Fix nan problem in precision_recall_curve (affecting average_precision_score) May 30, 2019
• 11 new comments Open #13902 [MRG] Adds fetch_openml pandas dataframe support May 31, 2019
• 10 new comments Open #12841 [MRG] Implement randomized PCA May 30, 2019
• 8 new comments Open #13908 [MRG] Fix formula of the objective function in BayesianRidge Jun 3, 2019
• 5 new comments Open #13603 [WIP] Add _n_features_in attribute to BaseEstimator May 31, 2019
• 5 new comments Open #13761 [MRG] Cross-validation for time series (inserting gaps between the training set and the test set) May 30, 2019
• 5 new comments Open #13817 [MRG] DEP change default validate in FunctionTransformer Jun 3, 2019
• 5 new comments Open #13875 [MRG] Binary only classification checks with estimator_checks Jun 2, 2019
• 4 new comments Open #13605 Gaussian Process kernel WhiteKernel has unhelpful "noise_level" documentation May 30, 2019
• 4 new comments Open #13900 [MRG] Successive halving for faster parameter search May 30, 2019
• 3 new comments Open #13783 knn predict unreasonably slow b/c of use of scipy.stats.mode May 28, 2019
• 3 new comments Open #9555 [WIP] Implement Gower similarity coeficient May 30, 2019
• 3 new comments Open #12887 [MRG] Adds Minimal Cost-Complexity Pruning to Decision Trees May 31, 2019
• 3 new comments Open #13204 [MRG] Feature: Additional `TimeSeriesSplit` Functionality May 30, 2019
• 3 new comments Open #13912 [NOMRG] PYPY Fix: try gc.collect after every run May 31, 2019
• 3 new comments Open #13925 [MRG] add _pairwise property to BaseSearchCV May 28, 2019
• 3 new comments Open #13935 [WIP] decision_path and apply for all tree-based ensemble model classes. Jun 1, 2019
• 3 new comments Open #13942 [MRG] Adds Bernoulli mixture model to the `mixture` module May 30, 2019
• 2 new comments Open #13940 Import Error when importing from different threads May 28, 2019
• 2 new comments Open #13949 Automatically publish to Releases section of repo Jun 3, 2019
• 2 new comments Open #12730 Add paired test for cross-validation Jun 3, 2019
• 2 new comments Open #13311 [MRG] Adds Deprecating Position Arguments Helper May 31, 2019
• 2 new comments Open #13645 MAINT: adjustments to test_logistic.py::test_dtype_match May 31, 2019
• 2 new comments Open #13914 add test to check that all ridge solver give the same results May 28, 2019
• 2 new comments Open #13939 [WIP] Bug fix for sklearn.metrics.cluster.normalized_mutual_info_score with sparse input May 30, 2019
• 1 new comment Open #12546 Parallalism in gridsearcCV is ending up with a permission error May 27, 2019
• 1 new comment Open #5318 types in dataset bunches are inconsistent May 28, 2019
• 1 new comment Open #12392 Target column names in datasets provided by Sklearn May 28, 2019
• 1 new comment Open #13773 IterativeImputer behaviour on missing nan's in fit data May 29, 2019
• 1 new comment Open #8245 average_precision_score does not return correct AP when all negative ground truth labels May 29, 2019
• 1 new comment Open #5884 Implement Gower Similarity Coefficient May 30, 2019
• 1 new comment Open #12957 Naive Bayes Classifier with Mixed Bernoulli/Gaussian Models May 31, 2019
• 1 new comment Open #13668 Support for Pandas DataFrames in kernels & heterogeneous kernel Jun 3, 2019
• 1 new comment Open #10482 [MRG] Generalize the use of precomputed sparse distance matrices, for estimators which use nearest neighbors May 30, 2019
• 1 new comment Open #13307 Feature names with input features May 31, 2019
• 1 new comment Open #13511 [MRG] Libsvm and liblinear rand() fix for convergence on windows targets May 29, 2019
• 1 new comment Open #13709 Bonify precision recall fscore support May 30, 2019
• 1 new comment Open #13930 [MRG] Fixes test_scale_and_stability by reducing the decimal May 31, 2019
• 0 new comments Open #13797 Remove deprecated functions and classes 0.22 May 29, 2019
• 0 new comments Open #12627 RFC Implement Pipeline get feature names May 31, 2019
• 0 new comments Open #13041 [MRG] Uses pytest-xdist to parallelize tests May 31, 2019