# Release history¶

## Version 0.18¶

### Changelog¶

#### New features¶

- The Gaussian Process module has been reimplemented and now offers classification and regression estimators through
gaussian_process.GaussianProcessClassifierandgaussian_process.GaussianProcessRegressor. Among other things, the new implementation supports kernel engineering, gradient-based hyperparameter optimization or sampling of functions from GP prior and GP posterior. Extensive documentation and examples are provided. By Jan Hendrik Metzen.- Added the
ensemble.IsolationForestclass for anomaly detection based on random forests. By Nicolas Goix.

#### Enhancements¶

- The cross-validation iterators are replaced by cross-validation splitters available from
model_selection. These expose asplitmethod that takes in the data and yields a generator for the different splits. This change makes it possible to do nested cross-validation with ease, facilitated bymodel_selection.GridSearchCVand similar utilities. (#4294) by Raghav R V.- The random forest, extra trees and decision tree estimators now has a method
decision_pathwhich returns the decision path of samples in the tree. By Arnaud Joly

- The random forest, extra tree and decision tree estimators now has a method
decision_pathwhich returns the decision path of samples in the tree. By Arnaud Joly- A new example has been added unveling the decision tree structure. By Arnaud Joly
- Random forest, extra trees, decision trees and gradient boosting estimator accept the parameter
min_samples_splitandmin_samples_leafprovided as a percentage of the training samples. By yelite and Arnaud Joly

#### Bug fixes¶

RandomizedPCAdefault number of iterated_power is 2 instead of 3. This is a speed up with a minor precision decrease. (#5141) by Giorgio Patrini.randomized_svdperforms 2 power iterations by default, instead or 0. In practice this is often enough for obtaining a good approximation of the true eigenvalues/vectors in the presence of noise. (#5141) by Giorgio Patrini.randomized_range_finderis more numerically stable when many power iterations are requested, since it applies LU normalization by default. If n_iter<2 numerical issues are unlikely, thus no normalization is applied. Other normalization options are available: ‘none’, ‘LU’ and ‘QR’. (#5141) by Giorgio Patrini.- Fixed bug in
manifold.spectral_embeddingwhere diagonal of unnormalized Laplacian matrix was incorrectly set to 1. By Peter Fischer.- Fixed incorrect initialization of
utils.arpack.eigshon all occurrences. Affectscluster.SpectralBiclustering,decomposition.KernelPCA,manifold.LocallyLinearEmbedding, andmanifold.SpectralEmbedding. By Peter Fischer.- Random forest, extra trees, decision trees and gradient boosting won’t accept anymore
min_samples_split=1as at least 2 samples are required to split a decision tree node. By Arnaud Joly

### API changes summary¶

- The
cross_validation,grid_searchandlearning_curvehave been deprecated and the classes and functions have been reorganized into themodel_selectionmodule. (#4294) by Raghav R V.

## Version 0.17¶

### Changelog¶

#### New features¶

- All the Scaler classes but
RobustScalercan be fitted online by calling partial_fit. By Giorgio Patrini.- The new class
ensemble.VotingClassifierimplements a “majority rule” / “soft voting” ensemble classifier to combine estimators for classification. By Sebastian Raschka.- The new class
preprocessing.RobustScalerprovides an alternative topreprocessing.StandardScalerfor feature-wise centering and range normalization that is robust to outliers. By Thomas Unterthiner.- The new class
preprocessing.MaxAbsScalerprovides an alternative topreprocessing.MinMaxScalerfor feature-wise range normalization when the data is already centered or sparse. By Thomas Unterthiner.- The new class
preprocessing.FunctionTransformerturns a Python function into aPipeline-compatible transformer object. By Joe Jevnik.- The new classes
cross_validation.LabelKFoldandcross_validation.LabelShuffleSplitgenerate train-test folds, respectively similar tocross_validation.KFoldandcross_validation.ShuffleSplit, except that the folds are conditioned on a label array. By Brian McFee, Jean Kossaifi and Gilles Louppe.decomposition.LatentDirichletAllocationimplements the Latent Dirichlet Allocation topic model with online variational inference. By Chyi-Kwei Yau, with code based on an implementation by Matt Hoffman. (#3659)- The new solver
sagimplements a Stochastic Average Gradient descent and is available in bothlinear_model.LogisticRegressionandlinear_model.Ridge. This solver is very efficient for large datasets. By Danny Sullivan and Tom Dupre la Tour. (#4738)- The new solver
cdimplements a Coordinate Descent indecomposition.NMF. Previous solver based on Projected Gradient is still available setting new parametersolvertopg, but is deprecated and will be removed in 0.19, along withdecomposition.ProjectedGradientNMFand parameterssparseness,eta,betaandnls_max_iter. New parametersalphaandl1_ratiocontrol L1 and L2 regularization, andshuffleadds a shuffling step in thecdsolver. By Tom Dupre la Tour and Mathieu Blondel.IndexErrorbug #5495 when doing OVR(SVC(decision_function_shape=”ovr”)). Fixed by Elvis Dohmatob.

#### Enhancements¶

manifold.TSNEnow supports approximate optimization via the Barnes-Hut method, leading to much faster fitting. By Christopher Erick Moody. (#4025)cluster.mean_shift_.MeanShiftnow supports parallel execution, as implemented in themean_shiftfunction. By Martino Sorbaro.naive_bayes.GaussianNBnow supports fitting withsample_weights. By Jan Hendrik Metzen.dummy.DummyClassifiernow supports a prior fitting strategy. By Arnaud Joly.- Added a
fit_predictmethod formixture.GMMand subclasses. By Cory Lorenz.- Added the
metrics.label_ranking_lossmetric. By Arnaud Joly.- Added the
metrics.cohen_kappa_scoremetric.- Added a
warm_startconstructor parameter to the bagging ensemble models to increase the size of the ensemble. By Tim Head.- Added option to use multi-output regression metrics without averaging. By Konstantin Shmelkov and Michael Eickenberg.
- Added
stratifyoption tocross_validation.train_test_splitfor stratified splitting. By Miroslav Batchkarov.- The
tree.export_graphvizfunction now supports aesthetic improvements fortree.DecisionTreeClassifierandtree.DecisionTreeRegressor, including options for coloring nodes by their majority class or impurity, showing variable names, and using node proportions instead of raw sample counts. By Trevor Stephens.- Improved speed of
newton-cgsolver inlinear_model.LogisticRegression, by avoiding loss computation. By Mathieu Blondel and Tom Dupre la Tour.- The
class_weight="auto"heuristic in classifiers supportingclass_weightwas deprecated and replaced by theclass_weight="balanced"option, which has a simpler forumlar and interpretation. By Hanna Wallach and Andreas Müller.- Add
class_weightparameter to automatically weight samples by class frequency forlinear_model.PassiveAgressiveClassifier. By Trevor Stephens.- Added backlinks from the API reference pages to the user guide. By Andreas Müller.
- The
labelsparameter tosklearn.metrics.f1_score,sklearn.metrics.fbeta_score,sklearn.metrics.recall_scoreandsklearn.metrics.precision_scorehas been extended. It is now possible to ignore one or more labels, such as where a multiclass problem has a majority class to ignore. By Joel Nothman.- Add
sample_weightsupport tolinear_model.RidgeClassifier. By Trevor Stephens.- Provide an option for sparse output from
sklearn.metrics.pairwise.cosine_similarity. By Jaidev Deshpande.- Add
minmax_scaleto provide a function interface forMinMaxScaler. By Thomas Unterthiner.dump_svmlight_filenow handles multi-label datasets. By Chih-Wei Chang.- RCV1 dataset loader (
sklearn.datasets.fetch_rcv1). By Tom Dupre la Tour.- The “Wisconsin Breast Cancer” classical two-class classification dataset is now included in scikit-learn, available with :fun:`sklearn.dataset.load_breast_cancer`.
- Upgraded to joblib 0.9.3 to benefit from the new automatic batching of short tasks. This makes it possible for scikit-learn to benefit from parallelism when many very short tasks are executed in parallel, for instance by the
grid_search.GridSearchCVmeta-estimator withn_jobs > 1used with a large grid of parameters on a small dataset. By Vlad Niculae, Olivier Grisel and Loic Esteve.- For more details about changes in joblib 0.9.3 see the release notes: https://github.com/joblib/joblib/blob/master/CHANGES.rst#release-093
- Improved speed (3 times per iteration) of
decomposition.DictLearningwith coordinate descent method fromlinear_model.Lasso. By Arthur Mensch.- Parallel processing (threaded) for queries of nearest neighbors (using the ball-tree) by Nikolay Mayorov.
- Allow
datasets.make_multilabel_classificationto output a sparsey. By Kashif Rasul.cluster.DBSCANnow accepts a sparse matrix of precomputed distances, allowing memory-efficient distance precomputation. By Joel Nothman.tree.DecisionTreeClassifiernow exposes anapplymethod for retrieving the leaf indices samples are predicted as. By Daniel Galvez and Gilles Louppe.- Speed up decision tree regressors, random forest regressors, extra trees regressors and gradient boosting estimators by computing a proxy of the impurity improvement during the tree growth. The proxy quantity is such that the split that maximizes this value also maximizes the impurity improvement. By Arnaud Joly, Jacob Schreiber and Gilles Louppe.
- Speed up tree based methods by reducing the number of computations needed when computing the impurity measure taking into account linear relationship of the computed statistics. The effect is particularly visible with extra trees and on datasets with categorical or sparse features. By Arnaud Joly.
ensemble.GradientBoostingRegressorandensemble.GradientBoostingClassifiernow expose anapplymethod for retrieving the leaf indices each sample ends up in under each try. By Jacob Schreiber.- Add
sample_weightsupport tolinear_model.LinearRegression. By Sonny Hu. (#4481)- Add
n_iter_without_progresstomanifold.TSNEto control the stopping criterion. By Santi Villalba. (#5185)- Added optional parameter
random_stateinlinear_model.Ridge, to set the seed of the pseudo random generator used insagsolver. By Tom Dupre la Tour.- Added optional parameter
warm_startinlinear_model.LogisticRegression. If set to True, the solverslbfgs,newton-cgandsagwill be initialized with the coefficients computed in the previous fit. By Tom Dupre la Tour.- Added
sample_weightsupport tolinear_model.LogisticRegressionfor thelbfgs,newton-cg, andsagsolvers. By Valentin Stolbunov. Support added to theliblinearsolver. By Manoj Kumar.- Added optional parameter
presorttoensemble.GradientBoostingRegressorandensemble.GradientBoostingClassifier, keeping default behavior the same. This allows gradient boosters to turn off presorting when building deep trees or using sparse data. By Jacob Schreiber.- Altered
metrics.roc_curveto drop unnecessary thresholds by default. By Graham Clenaghan.- Added
feature_selection.SelectFromModelmeta-transformer which can be used along with estimators that have coef_ or feature_importances_ attribute to select important features of the input data. By Maheshakya Wijewardena, Joel Nothman and Manoj Kumar.- Added
metrics.pairwise.laplacian_kernel. By Clyde Fare.covariance.GraphLassoallows separate control of the convergence criterion for the Elastic-Net subproblem via theenet_tolparameter.- Improved verbosity in
decomposition.DictionaryLearning.ensemble.RandomForestClassifierandensemble.RandomForestRegressorno longer explicitly store the samples used in bagging, resulting in a much reduced memory footprint for storing random forest models.- Added
positiveoption tolinear_model.Larsandlinear_model.lars_pathto force coefficients to be positive. (#5131 <https://github.com/scikit-learn/scikit-learn/pull/5131>)- Added the
X_norm_squaredparameter tometrics.pairwise.euclidean_distancesto provide precomputed squared norms forX.- Added the
fit_predictmethod topipeline.Pipeline.- Added the
preprocessing.min_max_scalefunction.

#### Bug fixes¶

- Fixed non-determinism in
dummy.DummyClassifierwith sparse multi-label output. By Andreas Müller.- Fixed the output shape of
linear_model.RANSACRegressorto(n_samples, ). By Andreas Müller.- Fixed bug in
decomposition.DictLearningwhenn_jobs < 0. By Andreas Müller.- Fixed bug where
grid_search.RandomizedSearchCVcould consume a lot of memory for large discrete grids. By Joel Nothman.- Fixed bug in
linear_model.LogisticRegressionCVwhere penalty was ignored in the final fit. By Manoj Kumar.- Fixed bug in
ensemble.forest.ForestClassifierwhile computing oob_score and X is a sparse.csc_matrix. By Ankur Ankan.- All regressors now consistently handle and warn when given
ythat is of shape(n_samples, 1). By Andreas Müller.- Fix in
cluster.KMeanscluster reassignment for sparse input by Lars Buitinck.- Fixed a bug in
lda.LDAthat could cause asymmetric covariance matrices when using shrinkage. By Martin Billinger.- Fixed
cross_validation.cross_val_predictfor estimators with sparse predictions. By Buddha Prakash.- Fixed the
predict_probamethod oflinear_model.LogisticRegressionto use soft-max instead of one-vs-rest normalization. By Manoj Kumar. (#5182)- Fixed the
partial_fitmethod oflinear_model.SGDClassifierwhen called withaverage=True. By Andrew Lamb. (#5282)- Dataset fetchers use different filenames under Python 2 and Python 3 to avoid pickling compatibility issues. By Olivier Grisel. (#5355)
- Fixed a bug in
naive_bayes.GaussianNBwhich caused classification results to depend on scale. By Jake Vanderplas.- Fixed temporarily
linear_model.Ridge, which was incorrect when fitting the intercept in the case of sparse data. The fix automatically changes the solver to ‘sag’ in this case. (#5360) By Tom Dupre la Tour.- Fixed a performance bug in
decomposition.RandomizedPCAon data with a large number of features and fewer samples. (#4478) By Andreas Müller, Loic Esteve and Giorgio Patrini.- Fixed bug in
cross_decomposition.PLSthat yielded unstable and platform dependent output, and failed on fit_transform. By Arthur Mensch.- Fixes to the
Bunchclass used to store datasets.- Fixed
ensemble.plot_partial_dependenceignoring thepercentilesparameter.- Providing a
setas vocabulary inCountVectorizerno longer leads to inconsistent results when pickling.- Fixed the conditions on when a precomputed Gram matrix needs to be recomputed in
linear_model.LinearRegression,linear_model.OrthogonalMatchingPursuit,linear_model.Lassoandlinear_model.ElasticNet.- Fixed inconsistent memory layout in the coordinate descent solver that affected
linear_model.DictionaryLearningandcovariance.GraphLasso. (#5337 <https://github.com/scikit-learn/scikit-learn/pull/5337>) By `Oliver Grisel`_.manifold.LocallyLinearEmbeddingno longer ignores theregparameter.- Nearest Neighbor estimators with custom distance metrics can now be pickled. (4362 <https://github.com/scikit-learn/scikit-learn/pull/4362>)
- Fixed a bug in
pipeline.FeatureUnionwheretransformer_weightswere not properly handled when performing grid-searches.- Fixed a bug in
linear_model.LogisticRegressionandlinear_model.LogisticRegressionCVwhen usingclass_weight='balanced'```or ``class_weight='auto'. By Tom Dupre la Tour.

### API changes summary¶

- Attribute data_min, data_max and data_range in
preprocessing.MinMaxScalerare deprecated and won’t be available from 0.19. Instead, the class now exposes data_min_, data_max_ and data_range_. By Giorgio Patrini.- All Scaler classes now have an scale_ attribute, the feature-wise rescaling applied by their transform methods. The old attribute std_ in
preprocessing.StandardScaleris deprecated and superseded by scale_; it won’t be available in 0.19. By Giorgio Patrini.svm.SVC`andsvm.NuSVCnow have andecision_function_shapeparameter to make their decision function of shape(n_samples, n_classes)by settingdecision_function_shape='ovr'. This will be the default behavior starting in 0.19. By Andreas Müller.- Passing 1D data arrays as input to estimators is now deprecated as it caused confusion in how the array elements should be interpreted as features or as samples. All data arrays are now expected to be explicitly shaped
(n_samples, n_features). By Vighnesh Birodkar.lda.LDAandqda.QDAhave been moved todiscriminant_analysis.LinearDiscriminantAnalysisanddiscriminant_analysis.QuadraticDiscriminantAnalysis.- The
store_covarianceandtolparameters have been moved from the fit method to the constructor indiscriminant_analysis.LinearDiscriminantAnalysisand thestore_covariancesandtolparameters have been moved from the fit method to the constructor indiscriminant_analysis.QuadraticDiscriminantAnalysis.- Models inheriting from
_LearntSelectorMixinwill no longer support the transform methods. (i.e, RandomForests, GradientBoosting, LogisticRegression, DecisionTrees, SVMs and SGD related models). Wrap these models around the metatransfomerfeature_selection.SelectFromModelto remove features (according to coefs_ or feature_importances_) which are below a certain threshold value instead.cluster.KMeansre-runs cluster-assignments in case of non-convergence, to ensure consistency ofpredict(X)andlabels_. By Vighnesh Birodkar.- Classifier and Regressor models are now tagged as such using the
_estimator_typeattribute.- Cross-validation iterators allways provide indices into training and test set, not boolean masks.
- The
decision_functionon all regressors was deprecated and will be removed in 0.19. Usepredictinstead.datasets.load_lfw_pairsis deprecated and will be removed in 0.19. Usedatasets.fetch_lfw_pairsinstead.- The deprecated
hmmmodule was removed.- The deprecated
Bootstrapcross-validation iterator was removed.- The deprecated
WardandWardAgglomerativeclasses have been removed. Useclustering.AgglomerativeClusteringinstead.cross_validation.check_cvis now a public function.- The property
residues_oflinear_model.LinearRegressionis deprecated and will be removed in 0.19.- The deprecated
n_jobsparameter oflinear_model.LinearRegressionhas been moved to the constructor.- Removed deprecated
class_weightparameter fromlinear_model.SGDClassifier‘sfitmethod. Use the construction parameter instead.- The deprecated support for the sequence of sequences (or list of lists) multilabel format was removed. To convert to and from the supported binary indicator matrix format, use
MultiLabelBinarizer.- The behavior of calling the
inverse_transformmethod ofPipeline.pipelinewill change in 0.19. It will no longer reshape one-dimensional input to two-dimensional input.- The deprecated attributes
indicator_matrix_,multilabel_andclasses_ofpreprocessing.LabelBinarizerwere removed.- Using
gamma=0insvm.SVCandsvm.SVRto automatically set the gamma to1. / n_featuresis deprecated and will be removed in 0.19. Usegamma="auto"instead.

## Version 0.16.1¶

### Changelog¶

#### Bug fixes¶

- Allow input data larger than
block_sizeincovariance.LedoitWolfby Andreas Müller.- Fix a bug in
isotonic.IsotonicRegressiondeduplication that caused unstable result incalibration.CalibratedClassifierCVby Jan Hendrik Metzen.- Fix sorting of labels in func:preprocessing.label_binarize by Michael Heilman.
- Fix several stability and convergence issues in
cross_decomposition.CCAandcross_decomposition.PLSCanonicalby Andreas Müller- Fix a bug in
cluster.KMeanswhenprecompute_distances=Falseon fortran-ordered data.- Fix a speed regression in
ensemble.RandomForestClassifier‘spredictandpredict_probaby Andreas Müller.- Fix a regression where
utils.shuffleconverted lists and dataframes to arrays, by Olivier Grisel

## Version 0.16¶

### Highlights¶

- Speed improvements (notably in
cluster.DBSCAN), reduced memory requirements, bug-fixes and better default settings.- Multinomial Logistic regression and a path algorithm in
linear_model.LogisticRegressionCV.- Out-of core learning of PCA via
decomposition.IncrementalPCA.- Probability callibration of classifiers using
calibration.CalibratedClassifierCV.cluster.Birchclustering method for large-scale datasets.- Scalable approximate nearest neighbors search with Locality-sensitive hashing forests in
neighbors.LSHForest.- Improved error messages and better validation when using malformed input data.
- More robust integration with pandas dataframes.

### Changelog¶

#### New features¶

- The new
neighbors.LSHForestimplements locality-sensitive hashing for approximate nearest neighbors search. By Maheshakya Wijewardena.- Added
svm.LinearSVR. This class uses the liblinear implementation of Support Vector Regression which is much faster for large sample sizes thansvm.SVRwith linear kernel. By Fabian Pedregosa and Qiang Luo.- Incremental fit for
GaussianNB.- Added
sample_weightsupport todummy.DummyClassifieranddummy.DummyRegressor. By Arnaud Joly.- Added the
metrics.label_ranking_average_precision_scoremetrics. By Arnaud Joly.- Add the
metrics.coverage_errormetrics. By Arnaud Joly.- Added
linear_model.LogisticRegressionCV. By Manoj Kumar, Fabian Pedregosa, Gael Varoquaux and Alexandre Gramfort.- Added
warm_startconstructor parameter to make it possible for any trained forest model to grow additional trees incrementally. By Laurent Direr.- Added
sample_weightsupport toensemble.GradientBoostingClassifierandensemble.GradientBoostingRegressor. By Peter Prettenhofer.- Added
decomposition.IncrementalPCA, an implementation of the PCA algorithm that supports out-of-core learning with apartial_fitmethod. By Kyle Kastner.- Averaged SGD for
SGDClassifierandSGDRegressorBy Danny Sullivan.- Added
cross_val_predictfunction which computes cross-validated estimates. By Luis Pedro Coelho- Added
linear_model.TheilSenRegressor, a robust generalized-median-based estimator. By Florian Wilhelm.- Added
metrics.median_absolute_error, a robust metric. By Gael Varoquaux and Florian Wilhelm.- Add
cluster.Birch, an online clustering algorithm. By Manoj Kumar, Alexandre Gramfort and Joel Nothman.- Added shrinkage support to
discriminant_analysis.LinearDiscriminantAnalysisusing two new solvers. By Clemens Brunner and Martin Billinger.- Added
kernel_ridge.KernelRidge, an implementation of kernelized ridge regression. By Mathieu Blondel and Jan Hendrik Metzen.- All solvers in
linear_model.Ridgenow support sample_weight. By Mathieu Blondel.- Added
cross_validation.PredefinedSplitcross-validation for fixed user-provided cross-validation folds. By Thomas Unterthiner.- Added
calibration.CalibratedClassifierCV, an approach for calibrating the predicted probabilities of a classifier. By Alexandre Gramfort, Jan Hendrik Metzen, Mathieu Blondel and Balazs Kegl.

#### Enhancements¶

- Add option
return_distanceinhierarchical.ward_treeto return distances between nodes for both structured and unstructured versions of the algorithm. By Matteo Visconti di Oleggio Castello. The same option was added inhierarchical.linkage_tree. By Manoj Kumar- Add support for sample weights in scorer objects. Metrics with sample weight support will automatically benefit from it. By Noel Dawe and Vlad Niculae.
- Added
newton-cgand lbfgs solver support inlinear_model.LogisticRegression. By Manoj Kumar.- Add
selection="random"parameter to implement stochastic coordinate descent forlinear_model.Lasso,linear_model.ElasticNetand related. By Manoj Kumar.- Add
sample_weightparameter tometrics.jaccard_similarity_scoreandmetrics.log_loss. By Jatin Shah.- Support sparse multilabel indicator representation in
preprocessing.LabelBinarizerandmulticlass.OneVsRestClassifier(by Hamzeh Alsalhi with thanks to Rohit Sivaprasad), as well as evaluation metrics (by Joel Nothman).- Add
sample_weightparameter to metrics.jaccard_similarity_score. By Jatin Shah.- Add support for multiclass in metrics.hinge_loss. Added
labels=Noneas optional paramter. By Saurabh Jha.- Add
sample_weightparameter to metrics.hinge_loss. By Saurabh Jha.- Add
multi_class="multinomial"option inlinear_model.LogisticRegressionto implement a Logistic Regression solver that minimizes the cross-entropy or multinomial loss instead of the default One-vs-Rest setting. Supports lbfgs and newton-cg solvers. By Lars Buitinck and Manoj Kumar. Solver option newton-cg by Simon Wu.DictVectorizercan now performfit_transformon an iterable in a single pass, when giving the optionsort=False. By Dan Blanchard.GridSearchCVandRandomizedSearchCVcan now be configured to work with estimators that may fail and raise errors on individual folds. This option is controlled by the error_score parameter. This does not affect errors raised on re-fit. By Michal Romaniuk.- Add
digitsparameter to metrics.classification_report to allow report to show different precision of floating point numbers. By Ian Gilmore.- Add a quantile prediction strategy to the
dummy.DummyRegressor. By Aaron Staple.- Add
handle_unknownoption topreprocessing.OneHotEncoderto handle unknown categorical features more gracefully during transform. By Manoj Kumar.- Added support for sparse input data to decision trees and their ensembles. By Fares Hedyati and Arnaud Joly.
- Optimized
cluster.AffinityPropagationby reducing the number of memory allocations of large temporary data-structures. By Antony Lee.- Parellization of the computation of feature importances in random forest. By Olivier Grisel and Arnaud Joly.
- Add
n_iter_attribute to estimators that accept amax_iterattribute in their constructor. By Manoj Kumar.- Added decision function for
multiclass.OneVsOneClassifierBy Raghav R V and Kyle Beauchamp.neighbors.kneighbors_graphandradius_neighbors_graphsupport non-Euclidean metrics. By Manoj Kumar- Parameter
connectivityincluster.AgglomerativeClusteringand family now accept callables that return a connectivity matrix. By Manoj Kumar.- Sparse support for
paired_distances. By Joel Nothman.cluster.DBSCANnow supports sparse input and sample weights and has been optimized: the inner loop has been rewritten in Cython and radius neighbors queries are now computed in batch. By Joel Nothman and Lars Buitinck.- Add
class_weightparameter to automatically weight samples by class frequency forensemble.RandomForestClassifier,tree.DecisionTreeClassifier,ensemble.ExtraTreesClassifierandtree.ExtraTreeClassifier. By Trevor Stephens.grid_search.RandomizedSearchCVnow does sampling without replacement if all parameters are given as lists. By Andreas Müller.- Parallelized calculation of
pairwise_distancesis now supported for scipy metrics and custom callables. By Joel Nothman.- Allow the fitting and scoring of all clustering algorithms in
pipeline.Pipeline. By Andreas Müller.- More robust seeding and improved error messages in
cluster.MeanShiftby Andreas Müller.- Make the stopping criterion for
mixture.GMM,mixture.DPGMMandmixture.VBGMMless dependent on the number of samples by thresholding the average log-likelihood change instead of its sum over all samples. By Hervé Bredin.- The outcome of
manifold.spectral_embeddingwas made deterministic by flipping the sign of eigen vectors. By Hasil Sharma.- Significant performance and memory usage improvements in
preprocessing.PolynomialFeatures. By Eric Martin.- Numerical stability improvements for
preprocessing.StandardScalerandpreprocessing.scale. By Nicolas Goixsvm.SVCfitted on sparse input now implementsdecision_function. By Rob Zinkov and Andreas Müller.cross_validation.train_test_splitnow preserves the input type, instead of converting to numpy arrays.

#### Documentation improvements¶

- Added example of using
FeatureUnionfor heterogeneous input. By Matt Terry- Documentation on scorers was improved, to highlight the handling of loss functions. By Matt Pico.
- A discrepancy between liblinear output and scikit-learn’s wrappers is now noted. By Manoj Kumar.
- Improved documentation generation: examples referring to a class or function are now shown in a gallery on the class/function’s API reference page. By Joel Nothman.
- More explicit documentation of sample generators and of data transformation. By Joel Nothman.
sklearn.neighbors.BallTreeandsklearn.neighbors.KDTreeused to point to empty pages stating that they are aliases of BinaryTree. This has been fixed to show the correct class docs. By Manoj Kumar.- Added silhouette plots for analysis of KMeans clustering using
metrics.silhouette_samplesandmetrics.silhouette_score. SeeSelecting the number of clusters with silhouette analysis on KMeans clustering

#### Bug fixes¶

- Metaestimators now support ducktyping for the presence of
decision_function,predict_probaand other methods. This fixes behavior ofgrid_search.GridSearchCV,grid_search.RandomizedSearchCV,pipeline.Pipeline,feature_selection.RFE,feature_selection.RFECVwhen nested. By Joel Nothman- The
scoringattribute of grid-search and cross-validation methods is no longer ignored when agrid_search.GridSearchCVis given as a base estimator or the base estimator doesn’t have predict.- The function
hierarchical.ward_treenow returns the children in the same order for both the structured and unstructured versions. By Matteo Visconti di Oleggio Castello.feature_selection.RFECVnow correctly handles cases whenstepis not equal to 1. By Nikolay Mayorov- The
decomposition.PCAnow undoes whitening in itsinverse_transform. Also, itscomponents_now always have unit length. By Michael Eickenberg.- Fix incomplete download of the dataset when
datasets.download_20newsgroupsis called. By Manoj Kumar.- Various fixes to the Gaussian processes subpackage by Vincent Dubourg and Jan Hendrik Metzen.
- Calling
partial_fitwithclass_weight=='auto'throws an appropriate error message and suggests a work around. By Danny Sullivan.RBFSamplerwithgamma=gformerly approximatedrbf_kernelwithgamma=g/2.; the definition ofgammais now consistent, which may substantially change your results if you use a fixed value. (If you cross-validated overgamma, it probably doesn’t matter too much.) By Dougal Sutherland.- Pipeline object delegate the
classes_attribute to the underlying estimator. It allows for instance to make bagging of a pipeline object. By Arnaud Jolyneighbors.NearestCentroidnow uses the median as the centroid when metric is set tomanhattan. It was using the mean before. By Manoj Kumar- Fix numerical stability issues in
linear_model.SGDClassifierandlinear_model.SGDRegressorby clipping large gradients and ensuring that weight decay rescaling is always positive (for large l2 regularization and large learning rate values). By Olivier Grisel- When compute_full_tree is set to “auto”, the full tree is built when n_clusters is high and is early stopped when n_clusters is low, while the behavior should be vice-versa in
cluster.AgglomerativeClustering(and friends). This has been fixed By Manoj Kumar- Fix lazy centering of data in
linear_model.enet_pathandlinear_model.lasso_path. It was centered around one. It has been changed to be centered around the origin. By Manoj Kumar- Fix handling of precomputed affinity matrices in
cluster.AgglomerativeClusteringwhen using connectivity constraints. By Cathy Deng- Correct
partial_fithandling ofclass_priorforsklearn.naive_bayes.MultinomialNBandsklearn.naive_bayes.BernoulliNB. By Trevor Stephens.- Fixed a crash in
metrics.precision_recall_fscore_supportwhen using unsortedlabelsin the multi-label setting. By Andreas Müller.- Avoid skipping the first nearest neighbor in the methods
radius_neighbors,kneighbors,kneighbors_graphandradius_neighbors_graphinsklearn.neighbors.NearestNeighborsand family, when the query data is not the same as fit data. By Manoj Kumar.- Fix log-density calculation in the
mixture.GMMwith tied covariance. By Will Dawson- Fixed a scaling error in
feature_selection.SelectFdrwhere a factorn_featureswas missing. By Andrew Tulloch- Fix zero division in
neighbors.KNeighborsRegressorand related classes when using distance weighting and having identical data points. By Garret-R.- Fixed round off errors with non positive-definite covariance matrices in GMM. By Alexis Mignon.
- Fixed a error in the computation of conditional probabilities in
naive_bayes.BernoulliNB. By Hanna Wallach.- Make the method
radius_neighborsofneighbors.NearestNeighborsreturn the samples lying on the boundary foralgorithm='brute'. By Yan Yi.- Flip sign of
dual_coef_ofsvm.SVCto make it consistent with the documentation anddecision_function. By Artem Sobolev.- Fixed handling of ties in
isotonic.IsotonicRegression. We now use the weighted average of targets (secondary method). By Andreas Müller and Michael Bommarito.

### API changes summary¶

GridSearchCVandcross_val_scoreand other meta-estimators don’t convert pandas DataFrames into arrays any more, allowing DataFrame specific operations in custom estimators.

multiclass.fit_ovr,multiclass.predict_ovr,predict_proba_ovr,multiclass.fit_ovo,multiclass.predict_ovo,multiclass.fit_ecocandmulticlass.predict_ecocare deprecated. Use the underlying estimators instead.Nearest neighbors estimators used to take arbitrary keyword arguments and pass these to their distance metric. This will no longer be supported in scikit-learn 0.18; use the

metric_paramsargument instead.

- n_jobs parameter of the fit method shifted to the constructor of the
LinearRegression class.

The

predict_probamethod ofmulticlass.OneVsRestClassifiernow returns two probabilities per sample in the multiclass case; this is consistent with other estimators and with the method’s documentation, but previous versions accidentally returned only the positive probability. Fixed by Will Lamond and Lars Buitinck.Change default value of precompute in

ElasticNetandLassoto False. Setting precompute to “auto” was found to be slower when n_samples > n_features since the computation of the Gram matrix is computationally expensive and outweighs the benefit of fitting the Gram for just one alpha.precompute="auto"is now deprecated and will be removed in 0.18 By Manoj Kumar.Expose

positiveoption inlinear_model.enet_pathandlinear_model.enet_pathwhich constrains coefficients to be positive. By Manoj Kumar.Users should now supply an explicit

averageparameter tosklearn.metrics.f1_score,sklearn.metrics.fbeta_score,sklearn.metrics.recall_scoreandsklearn.metrics.precision_scorewhen performing multiclass or multilabel (i.e. not binary) classification. By Joel Nothman.scoring parameter for cross validation now accepts ‘f1_micro’, ‘f1_macro’ or ‘f1_weighted’. ‘f1’ is now for binary classification only. Similar changes apply to ‘precision’ and ‘recall’. By Joel Nothman.

The

fit_intercept,normalizeandreturn_modelsparameters inlinear_model.enet_pathandlinear_model.lasso_pathhave been removed. They were deprecated since 0.14From now onwards, all estimators will uniformly raise

NotFittedError(utils.validation.NotFittedError), when any of thepredictlike methods are called before the model is fit. By Raghav R V.Input data validation was refactored for more consistent input validation. The

check_arraysfunction was replaced bycheck_arrayandcheck_X_y. By Andreas Müller.Allow

X=Nonein the methodsradius_neighbors,kneighbors,kneighbors_graphandradius_neighbors_graphinsklearn.neighbors.NearestNeighborsand family. If set to None, then for every sample this avoids setting the sample itself as the first nearest neighbor. By Manoj Kumar.Add parameter

include_selfinneighbors.kneighbors_graphandneighbors.radius_neighbors_graphwhich has to be explicitly set by the user. If set to True, then the sample itself is considered as the first nearest neighbor.thresh parameter is deprecated in favor of new tol parameter in

GMM,DPGMMandVBGMM. See Enhancements section for details. By Hervé Bredin.Estimators will treat input with dtype object as numeric when possible. By Andreas Müller

Estimators now raise ValueError consistently when fitted on empty data (less than 1 sample or less than 1 feature for 2D input). By Olivier Grisel.

The

shuffleoption oflinear_model.SGDClassifier,linear_model.SGDRegressor,linear_model.Perceptron,linear_model.PassiveAgressiveClassifierandlinear_model.PassiveAgressiveRegressornow defaults toTrue.

cluster.DBSCANnow uses a deterministic initialization. The random_state parameter is deprecated. By Erich Schubert.

## Version 0.15.2¶

### Bug fixes¶

- Fixed handling of the
pparameter of the Minkowski distance that was previously ignored in nearest neighbors models. By Nikolay Mayorov.- Fixed duplicated alphas in
linear_model.LassoLarswith early stopping on 32 bit Python. By Olivier Grisel and Fabian Pedregosa.- Fixed the build under Windows when scikit-learn is built with MSVC while NumPy is built with MinGW. By Olivier Grisel and Federico Vaggi.
- Fixed an array index overflow bug in the coordinate descent solver. By Gael Varoquaux.
- Better handling of numpy 1.9 deprecation warnings. By Gael Varoquaux.
- Removed unnecessary data copy in
cluster.KMeans. By Gael Varoquaux.- Explicitly close open files to avoid
ResourceWarningsunder Python 3. By Calvin Giles.- The
transformofdiscriminant_analysis.LinearDiscriminantAnalysisnow projects the input on the most discriminant directions. By Martin Billinger.- Fixed potential overflow in
_tree.safe_reallocby Lars Buitinck.- Performance optimization in
isotonic.IsotonicRegression. By Robert Bradshaw.noseis non-longer a runtime dependency to importsklearn, only for running the tests. By Joel Nothman.- Many documentation and website fixes by Joel Nothman, Lars Buitinck Matt Pico, and others.

## Version 0.15.1¶

### Bug fixes¶

- Made
cross_validation.cross_val_scoreusecross_validation.KFoldinstead ofcross_validation.StratifiedKFoldon multi-output classification problems. By Nikolay Mayorov.- Support unseen labels
preprocessing.LabelBinarizerto restore the default behavior of 0.14.1 for backward compatibility. By Hamzeh Alsalhi.- Fixed the
cluster.KMeansstopping criterion that prevented early convergence detection. By Edward Raff and Gael Varoquaux.- Fixed the behavior of
multiclass.OneVsOneClassifier. in case of ties at the per-class vote level by computing the correct per-class sum of prediction scores. By Andreas Müller.- Made
cross_validation.cross_val_scoreandgrid_search.GridSearchCVaccept Python lists as input data. This is especially useful for cross-validation and model selection of text processing pipelines. By Andreas Müller.- Fixed data input checks of most estimators to accept input data that implements the NumPy
__array__protocol. This is the case for forpandas.Seriesandpandas.DataFramein recent versions of pandas. By Gael Varoquaux.- Fixed a regression for
linear_model.SGDClassifierwithclass_weight="auto"on data with non-contiguous labels. By Olivier Grisel.

## Version 0.15¶

### Highlights¶

- Many speed and memory improvements all across the code
- Huge speed and memory improvements to random forests (and extra trees) that also benefit better from parallel computing.
- Incremental fit to
BernoulliRBM- Added
cluster.AgglomerativeClusteringfor hierarchical agglomerative clustering with average linkage, complete linkage and ward strategies.- Added
linear_model.RANSACRegressorfor robust regression models.- Added dimensionality reduction with
manifold.TSNEwhich can be used to visualize high-dimensional data.

### Changelog¶

#### New features¶

- Added
ensemble.BaggingClassifierandensemble.BaggingRegressormeta-estimators for ensembling any kind of base estimator. See theBaggingsection of the user guide for details and examples. By Gilles Louppe.- New unsupervised feature selection algorithm
feature_selection.VarianceThreshold, by Lars Buitinck.- Added
linear_model.RANSACRegressormeta-estimator for the robust fitting of regression models. By Johannes Schönberger.- Added
cluster.AgglomerativeClusteringfor hierarchical agglomerative clustering with average linkage, complete linkage and ward strategies, by Nelle Varoquaux and Gael Varoquaux.- Shorthand constructors
pipeline.make_pipelineandpipeline.make_unionwere added by Lars Buitinck.- Shuffle option for
cross_validation.StratifiedKFold. By Jeffrey Blackburne.- Incremental learning (
partial_fit) for Gaussian Naive Bayes by Imran Haque.- Added
partial_fittoBernoulliRBMBy Danny Sullivan.- Added
learning_curveutility to chart performance with respect to training size. SeePlotting Learning Curves. By Alexander Fabisch.- Add positive option in
LassoCVandElasticNetCV. By Brian Wignall and Alexandre Gramfort.- Added
linear_model.MultiTaskElasticNetCVandlinear_model.MultiTaskLassoCV. By Manoj Kumar.- Added
manifold.TSNE. By Alexander Fabisch.

#### Enhancements¶

- Add sparse input support to
ensemble.AdaBoostClassifierandensemble.AdaBoostRegressormeta-estimators. By Hamzeh Alsalhi.- Memory improvements of decision trees, by Arnaud Joly.
- Decision trees can now be built in best-first manner by using
max_leaf_nodesas the stopping criteria. Refactored the tree code to use either a stack or a priority queue for tree building. By Peter Prettenhofer and Gilles Louppe.- Decision trees can now be fitted on fortran- and c-style arrays, and non-continuous arrays without the need to make a copy. If the input array has a different dtype than
np.float32, a fortran- style copy will be made since fortran-style memory layout has speed advantages. By Peter Prettenhofer and Gilles Louppe.- Speed improvement of regression trees by optimizing the the computation of the mean square error criterion. This lead to speed improvement of the tree, forest and gradient boosting tree modules. By Arnaud Joly
- The
img_to_graphandgrid_tographfunctions insklearn.feature_extraction.imagenow returnnp.ndarrayinstead ofnp.matrixwhenreturn_as=np.ndarray. See the Notes section for more information on compatibility.- Changed the internal storage of decision trees to use a struct array. This fixed some small bugs, while improving code and providing a small speed gain. By Joel Nothman.
- Reduce memory usage and overhead when fitting and predicting with forests of randomized trees in parallel with
n_jobs != 1by leveraging new threading backend of joblib 0.8 and releasing the GIL in the tree fitting Cython code. By Olivier Grisel and Gilles Louppe.- Speed improvement of the
sklearn.ensemble.gradient_boostingmodule. By Gilles Louppe and Peter Prettenhofer.- Various enhancements to the
sklearn.ensemble.gradient_boostingmodule: awarm_startargument to fit additional trees, amax_leaf_nodesargument to fit GBM style trees, amonitorfit argument to inspect the estimator during training, and refactoring of the verbose code. By Peter Prettenhofer.- Faster
sklearn.ensemble.ExtraTreesby caching feature values. By Arnaud Joly.- Faster depth-based tree building algorithm such as decision tree, random forest, extra trees or gradient tree boosting (with depth based growing strategy) by avoiding trying to split on found constant features in the sample subset. By Arnaud Joly.
- Add
min_weight_fraction_leafpre-pruning parameter to tree-based methods: the minimum weighted fraction of the input samples required to be at a leaf node. By Noel Dawe.- Added
metrics.pairwise_distances_argmin_min, by Philippe Gervais.- Added predict method to
cluster.AffinityPropagationandcluster.MeanShift, by Mathieu Blondel.- Vector and matrix multiplications have been optimised throughout the library by Denis Engemann, and Alexandre Gramfort. In particular, they should take less memory with older NumPy versions (prior to 1.7.2).
- Precision-recall and ROC examples now use train_test_split, and have more explanation of why these metrics are useful. By Kyle Kastner
- The training algorithm for
decomposition.NMFis faster for sparse matrices and has much lower memory complexity, meaning it will scale up gracefully to large datasets. By Lars Buitinck.- Added svd_method option with default value to “randomized” to
decomposition.FactorAnalysisto save memory and significantly speedup computation by Denis Engemann, and Alexandre Gramfort.- Changed
cross_validation.StratifiedKFoldto try and preserve as much of the original ordering of samples as possible so as not to hide overfitting on datasets with a non-negligible level of samples dependency. By Daniel Nouri and Olivier Grisel.- Add multi-output support to
gaussian_process.GaussianProcessby John Novak.- Support for precomputed distance matrices in nearest neighbor estimators by Robert Layton and Joel Nothman.
- Norm computations optimized for NumPy 1.6 and later versions by Lars Buitinck. In particular, the k-means algorithm no longer needs a temporary data structure the size of its input.
dummy.DummyClassifiercan now be used to predict a constant output value. By Manoj Kumar.dummy.DummyRegressorhas now a strategy parameter which allows to predict the mean, the median of the training set or a constant output value. By Maheshakya Wijewardena.- Multi-label classification output in multilabel indicator format is now supported by
metrics.roc_auc_scoreandmetrics.average_precision_scoreby Arnaud Joly.- Significant performance improvements (more than 100x speedup for large problems) in
isotonic.IsotonicRegressionby Andrew Tulloch.- Speed and memory usage improvements to the SGD algorithm for linear models: it now uses threads, not separate processes, when
n_jobs>1. By Lars Buitinck.- Grid search and cross validation allow NaNs in the input arrays so that preprocessors such as
preprocessing.Imputercan be trained within the cross validation loop, avoiding potentially skewed results.- Ridge regression can now deal with sample weights in feature space (only sample space until then). By Michael Eickenberg. Both solutions are provided by the Cholesky solver.
- Several classification and regression metrics now support weighted samples with the new
sample_weightargument:metrics.accuracy_score,metrics.zero_one_loss,metrics.precision_score,metrics.average_precision_score,metrics.f1_score,metrics.fbeta_score,metrics.recall_score,metrics.roc_auc_score,metrics.explained_variance_score,metrics.mean_squared_error,metrics.mean_absolute_error,metrics.r2_score. By Noel Dawe.- Speed up of the sample generator
datasets.make_multilabel_classification. By Joel Nothman.

#### Documentation improvements¶

- The
Working With Text Datatutorial has now been worked in to the main documentation’s tutorial section. Includes exercises and skeletons for tutorial presentation. Original tutorial created by several authors including Olivier Grisel, Lars Buitinck and many others. Tutorial integration into the scikit-learn documentation by Jaques Grobler- Added
Computational Performancedocumentation. Discussion and examples of prediction latency / throughput and different factors that have influence over speed. Additional tips for building faster models and choosing a relevant compromise between speed and predictive power. By Eustache Diemert.

#### Bug fixes¶

- Fixed bug in
decomposition.MiniBatchDictionaryLearning:partial_fitwas not working properly.- Fixed bug in
linear_model.stochastic_gradient:l1_ratiowas used as(1.0 - l1_ratio).- Fixed bug in
multiclass.OneVsOneClassifierwith string labels- Fixed a bug in
LassoCVandElasticNetCV: they would not pre-compute the Gram matrix withprecompute=Trueorprecompute="auto"andn_samples > n_features. By Manoj Kumar.- Fixed incorrect estimation of the degrees of freedom in
feature_selection.f_regressionwhen variates are not centered. By Virgile Fritsch.- Fixed a race condition in parallel processing with
pre_dispatch != "all"(for instance incross_val_score). By Olivier Grisel.- Raise error in
cluster.FeatureAgglomerationandcluster.WardAgglomerationwhen no samples are given, rather than returning meaningless clustering.- Fixed bug in
gradient_boosting.GradientBoostingRegressorwithloss='huber':gammamight have not been initialized.- Fixed feature importances as computed with a forest of randomized trees when fit with
sample_weight != Noneand/or withbootstrap=True. By Gilles Louppe.

### API changes summary¶

sklearn.hmmis deprecated. Its removal is planned for the 0.17 release.- Use of
covariance.EllipticEnvelophas now been removed after deprecation. Please usecovariance.EllipticEnvelopeinstead.cluster.Wardis deprecated. Usecluster.AgglomerativeClusteringinstead.cluster.WardClusteringis deprecated. Usecluster.AgglomerativeClusteringinstead.cross_validation.Bootstrapis deprecated.cross_validation.KFoldorcross_validation.ShuffleSplitare recommended instead.- Direct support for the sequence of sequences (or list of lists) multilabel format is deprecated. To convert to and from the supported binary indicator matrix format, use
MultiLabelBinarizer. By Joel Nothman.- Add score method to
PCAfollowing the model of probabilistic PCA and deprecateProbabilisticPCAmodel whose score implementation is not correct. The computation now also exploits the matrix inversion lemma for faster computation. By Alexandre Gramfort.- The score method of
FactorAnalysisnow returns the average log-likelihood of the samples. Use score_samples to get log-likelihood of each sample. By Alexandre Gramfort.- Generating boolean masks (the setting
indices=False) from cross-validation generators is deprecated. Support for masks will be removed in 0.17. The generators have produced arrays of indices by default since 0.10. By Joel Nothman.- 1-d arrays containing strings with
dtype=object(as used in Pandas) are now considered valid classification targets. This fixes a regression from version 0.13 in some classifiers. By Joel Nothman.- Fix wrong
explained_variance_ratio_attribute inRandomizedPCA. By Alexandre Gramfort.- Fit alphas for each
l1_ratioinstead ofmean_l1_ratioinlinear_model.ElasticNetCVandlinear_model.LassoCV. This changes the shape ofalphas_from(n_alphas,)to(n_l1_ratio, n_alphas)if thel1_ratioprovided is a 1-D array like object of length greater than one. By Manoj Kumar.- Fix
linear_model.ElasticNetCVandlinear_model.LassoCVwhen fitting intercept and input data is sparse. The automatic grid of alphas was not computed correctly and the scaling with normalize was wrong. By Manoj Kumar.- Fix wrong maximal number of features drawn (
max_features) at each split for decision trees, random forests and gradient tree boosting. Previously, the count for the number of drawn features started only after one non constant features in the split. This bug fix will affect computational and generalization performance of those algorithms in the presence of constant features. To get back previous generalization performance, you should modify the value ofmax_features. By Arnaud Joly.- Fix wrong maximal number of features drawn (
max_features) at each split forensemble.ExtraTreesClassifierandensemble.ExtraTreesRegressor. Previously, only non constant features in the split was counted as drawn. Now constant features are counted as drawn. Furthermore at least one feature must be non constant in order to make a valid split. This bug fix will affect computational and generalization performance of extra trees in the presence of constant features. To get back previous generalization performance, you should modify the value ofmax_features. By Arnaud Joly.- Fix
utils.compute_class_weightwhenclass_weight=="auto". Previously it was broken for input of non-integerdtypeand the weighted array that was returned was wrong. By Manoj Kumar.- Fix
cross_validation.Bootstrapto returnValueErrorwhenn_train + n_test > n. By Ronald Phlypo.

### People¶

List of contributors for release 0.15 by number of commits.

- 312 Olivier Grisel
- 275 Lars Buitinck
- 221 Gael Varoquaux
- 148 Arnaud Joly
- 134 Johannes Schönberger
- 119 Gilles Louppe
- 113 Joel Nothman
- 111 Alexandre Gramfort
- 95 Jaques Grobler
- 89 Denis Engemann
- 83 Peter Prettenhofer
- 83 Alexander Fabisch
- 62 Mathieu Blondel
- 60 Eustache Diemert
- 60 Nelle Varoquaux
- 49 Michael Bommarito
- 45 Manoj-Kumar-S
- 28 Kyle Kastner
- 26 Andreas Mueller
- 22 Noel Dawe
- 21 Maheshakya Wijewardena
- 21 Brooke Osborn
- 21 Hamzeh Alsalhi
- 21 Jake VanderPlas
- 21 Philippe Gervais
- 19 Bala Subrahmanyam Varanasi
- 12 Ronald Phlypo
- 10 Mikhail Korobov
- 8 Thomas Unterthiner
- 8 Jeffrey Blackburne
- 8 eltermann
- 8 bwignall
- 7 Ankit Agrawal
- 7 CJ Carey
- 6 Daniel Nouri
- 6 Chen Liu
- 6 Michael Eickenberg
- 6 ugurthemaster
- 5 Aaron Schumacher
- 5 Baptiste Lagarde
- 5 Rajat Khanduja
- 5 Robert McGibbon
- 5 Sergio Pascual
- 4 Alexis Metaireau
- 4 Ignacio Rossi
- 4 Virgile Fritsch
- 4 Sebastian Saeger
- 4 Ilambharathi Kanniah
- 4 sdenton4
- 4 Robert Layton
- 4 Alyssa
- 4 Amos Waterland
- 3 Andrew Tulloch
- 3 murad
- 3 Steven Maude
- 3 Karol Pysniak
- 3 Jacques Kvam
- 3 cgohlke
- 3 cjlin
- 3 Michael Becker
- 3 hamzeh
- 3 Eric Jacobsen
- 3 john collins
- 3 kaushik94
- 3 Erwin Marsi
- 2 csytracy
- 2 LK
- 2 Vlad Niculae
- 2 Laurent Direr
- 2 Erik Shilts
- 2 Raul Garreta
- 2 Yoshiki Vázquez Baeza
- 2 Yung Siang Liau
- 2 abhishek thakur
- 2 James Yu
- 2 Rohit Sivaprasad
- 2 Roland Szabo
- 2 amormachine
- 2 Alexis Mignon
- 2 Oscar Carlsson
- 2 Nantas Nardelli
- 2 jess010
- 2 kowalski87
- 2 Andrew Clegg
- 2 Federico Vaggi
- 2 Simon Frid
- 2 Félix-Antoine Fortin
- 1 Ralf Gommers
- 1 t-aft
- 1 Ronan Amicel
- 1 Rupesh Kumar Srivastava
- 1 Ryan Wang
- 1 Samuel Charron
- 1 Samuel St-Jean
- 1 Fabian Pedregosa
- 1 Skipper Seabold
- 1 Stefan Walk
- 1 Stefan van der Walt
- 1 Stephan Hoyer
- 1 Allen Riddell
- 1 Valentin Haenel
- 1 Vijay Ramesh
- 1 Will Myers
- 1 Yaroslav Halchenko
- 1 Yoni Ben-Meshulam
- 1 Yury V. Zaytsev
- 1 adrinjalali
- 1 ai8rahim
- 1 alemagnani
- 1 alex
- 1 benjamin wilson
- 1 chalmerlowe
- 1 dzikie drożdże
- 1 jamestwebber
- 1 matrixorz
- 1 popo
- 1 samuela
- 1 François Boulogne
- 1 Alexander Measure
- 1 Ethan White
- 1 Guilherme Trein
- 1 Hendrik Heuer
- 1 IvicaJovic
- 1 Jan Hendrik Metzen
- 1 Jean Michel Rouly
- 1 Eduardo Ariño de la Rubia
- 1 Jelle Zijlstra
- 1 Eddy L O Jansson
- 1 Denis
- 1 John
- 1 John Schmidt
- 1 Jorge Cañardo Alastuey
- 1 Joseph Perla
- 1 Joshua Vredevoogd
- 1 José Ricardo
- 1 Julien Miotte
- 1 Kemal Eren
- 1 Kenta Sato
- 1 David Cournapeau
- 1 Kyle Kelley
- 1 Daniele Medri
- 1 Laurent Luce
- 1 Laurent Pierron
- 1 Luis Pedro Coelho
- 1 DanielWeitzenfeld
- 1 Craig Thompson
- 1 Chyi-Kwei Yau
- 1 Matthew Brett
- 1 Matthias Feurer
- 1 Max Linke
- 1 Chris Filo Gorgolewski
- 1 Charles Earl
- 1 Michael Hanke
- 1 Michele Orrù
- 1 Bryan Lunt
- 1 Brian Kearns
- 1 Paul Butler
- 1 Paweł Mandera
- 1 Peter
- 1 Andrew Ash
- 1 Pietro Zambelli
- 1 staubda

## Version 0.14¶

### Changelog¶

- Missing values with sparse and dense matrices can be imputed with the transformer
preprocessing.Imputerby Nicolas Trésegnie.- The core implementation of decisions trees has been rewritten from scratch, allowing for faster tree induction and lower memory consumption in all tree-based estimators. By Gilles Louppe.
- Added
ensemble.AdaBoostClassifierandensemble.AdaBoostRegressor, by Noel Dawe and Gilles Louppe. See theAdaBoostsection of the user guide for details and examples.- Added
grid_search.RandomizedSearchCVandgrid_search.ParameterSamplerfor randomized hyperparameter optimization. By Andreas Müller.- Added
biclusteringalgorithms (sklearn.cluster.bicluster.SpectralCoclusteringandsklearn.cluster.bicluster.SpectralBiclustering), data generation methods (sklearn.datasets.make_biclustersandsklearn.datasets.make_checkerboard), and scoring metrics (sklearn.metrics.consensus_score). By Kemal Eren.- Added
Restricted Boltzmann Machines(neural_network.BernoulliRBM). By Yann Dauphin.- Python 3 support by Justin Vincent, Lars Buitinck, Subhodeep Moitra and Olivier Grisel. All tests now pass under Python 3.3.
- Ability to pass one penalty (alpha value) per target in
linear_model.Ridge, by @eickenberg and Mathieu Blondel.- Fixed
sklearn.linear_model.stochastic_gradient.pyL2 regularization issue (minor practical significance). By Norbert Crombach and Mathieu Blondel .- Added an interactive version of Andreas Müller‘s Machine Learning Cheat Sheet (for scikit-learn) to the documentation. See
Choosing the right estimator. By Jaques Grobler.grid_search.GridSearchCVandcross_validation.cross_val_scorenow support the use of advanced scoring function such as area under the ROC curve and f-beta scores. SeeThe scoring parameter: defining model evaluation rulesfor details. By Andreas Müller and Lars Buitinck. Passing a function fromsklearn.metricsasscore_funcis deprecated.- Multi-label classification output is now supported by
metrics.accuracy_score,metrics.zero_one_loss,metrics.f1_score,metrics.fbeta_score,metrics.classification_report,metrics.precision_scoreandmetrics.recall_scoreby Arnaud Joly.- Two new metrics
metrics.hamming_lossandmetrics.jaccard_similarity_scoreare added with multi-label support by Arnaud Joly.- Speed and memory usage improvements in
feature_extraction.text.CountVectorizerandfeature_extraction.text.TfidfVectorizer, by Jochen Wersdörfer and Roman Sinayev.- The
min_dfparameter infeature_extraction.text.CountVectorizerandfeature_extraction.text.TfidfVectorizer, which used to be 2, has been reset to 1 to avoid unpleasant surprises (empty vocabularies) for novice users who try it out on tiny document collections. A value of at least 2 is still recommended for practical use.svm.LinearSVC,linear_model.SGDClassifierandlinear_model.SGDRegressornow have asparsifymethod that converts theircoef_into a sparse matrix, meaning stored models trained using these estimators can be made much more compact.linear_model.SGDClassifiernow produces multiclass probability estimates when trained under log loss or modified Huber loss.- Hyperlinks to documentation in example code on the website by Martin Luessi.
- Fixed bug in
preprocessing.MinMaxScalercausing incorrect scaling of the features for non-defaultfeature_rangesettings. By Andreas Müller.max_featuresintree.DecisionTreeClassifier,tree.DecisionTreeRegressorand all derived ensemble estimators now supports percentage values. By Gilles Louppe.- Performance improvements in
isotonic.IsotonicRegressionby Nelle Varoquaux.metrics.accuracy_scorehas an option normalize to return the fraction or the number of correctly classified sample by Arnaud Joly.- Added
metrics.log_lossthat computes log loss, aka cross-entropy loss. By Jochen Wersdörfer and Lars Buitinck.- A bug that caused
ensemble.AdaBoostClassifier‘s to output incorrect probabilities has been fixed.- Feature selectors now share a mixin providing consistent
transform,inverse_transformandget_supportmethods. By Joel Nothman.- A fitted
grid_search.GridSearchCVorgrid_search.RandomizedSearchCVcan now generally be pickled. By Joel Nothman.- Refactored and vectorized implementation of
metrics.roc_curveandmetrics.precision_recall_curve. By Joel Nothman.- The new estimator
sklearn.decomposition.TruncatedSVDperforms dimensionality reduction using SVD on sparse matrices, and can be used for latent semantic analysis (LSA). By Lars Buitinck.- Added self-contained example of out-of-core learning on text data
Out-of-core classification of text documents. By Eustache Diemert.- The default number of components for
sklearn.decomposition.RandomizedPCAis now correctly documented to ben_features. This was the default behavior, so programs using it will continue to work as they did.sklearn.cluster.KMeansnow fits several orders of magnitude faster on sparse data (the speedup depends on the sparsity). By Lars Buitinck.- Reduce memory footprint of FastICA by Denis Engemann and Alexandre Gramfort.
- Verbose output in
sklearn.ensemble.gradient_boostingnow uses a column format and prints progress in decreasing frequency. It also shows the remaining time. By Peter Prettenhofer.sklearn.ensemble.gradient_boostingprovides out-of-bag improvementoob_improvement_rather than the OOB score for model selection. An example that shows how to use OOB estimates to select the number of trees was added. By Peter Prettenhofer.- Most metrics now support string labels for multiclass classification by Arnaud Joly and Lars Buitinck.
- New OrthogonalMatchingPursuitCV class by Alexandre Gramfort and Vlad Niculae.
- Fixed a bug in
sklearn.covariance.GraphLassoCV: the ‘alphas’ parameter now works as expected when given a list of values. By Philippe Gervais.- Fixed an important bug in
sklearn.covariance.GraphLassoCVthat prevented all folds provided by a CV object to be used (only the first 3 were used). When providing a CV object, execution time may thus increase significantly compared to the previous version (bug results are correct now). By Philippe Gervais.cross_validation.cross_val_scoreand thegrid_searchmodule is now tested with multi-output data by Arnaud Joly.datasets.make_multilabel_classificationcan now return the output in label indicator multilabel format by Arnaud Joly.- K-nearest neighbors,
neighbors.KNeighborsRegressorandneighbors.RadiusNeighborsRegressor, and radius neighbors,neighbors.RadiusNeighborsRegressorandneighbors.RadiusNeighborsClassifiersupport multioutput data by Arnaud Joly.- Random state in LibSVM-based estimators (
svm.SVC,NuSVC,OneClassSVM,svm.SVR,svm.NuSVR) can now be controlled. This is useful to ensure consistency in the probability estimates for the classifiers trained withprobability=True. By Vlad Niculae.- Out-of-core learning support for discrete naive Bayes classifiers
sklearn.naive_bayes.MultinomialNBandsklearn.naive_bayes.BernoulliNBby adding thepartial_fitmethod by Olivier Grisel.- New website design and navigation by Gilles Louppe, Nelle Varoquaux, Vincent Michel and Andreas Müller.
- Improved documentation on
multi-class, multi-label and multi-output classificationby Yannick Schwartz and Arnaud Joly.- Better input and error handling in the
metricsmodule by Arnaud Joly and Joel Nothman.- Speed optimization of the
hmmmodule by Mikhail Korobov- Significant speed improvements for
sklearn.cluster.DBSCANby cleverless

### API changes summary¶

- The
auc_scorewas renamedroc_auc_score.- Testing scikit-learn with
sklearn.test()is deprecated. Usenosetests sklearnfrom the command line.- Feature importances in
tree.DecisionTreeClassifier,tree.DecisionTreeRegressorand all derived ensemble estimators are now computed on the fly when accessing thefeature_importances_attribute. Settingcompute_importances=Trueis no longer required. By Gilles Louppe.linear_model.lasso_pathandlinear_model.enet_pathcan return its results in the same format as that oflinear_model.lars_path. This is done by setting thereturn_modelsparameter toFalse. By Jaques Grobler and Alexandre Gramfortgrid_search.IterGridwas renamed togrid_search.ParameterGrid.- Fixed bug in
KFoldcausing imperfect class balance in some cases. By Alexandre Gramfort and Tadej Janež.sklearn.neighbors.BallTreehas been refactored, and asklearn.neighbors.KDTreehas been added which shares the same interface. The Ball Tree now works with a wide variety of distance metrics. Both classes have many new methods, including single-tree and dual-tree queries, breadth-first and depth-first searching, and more advanced queries such as kernel density estimation and 2-point correlation functions. By Jake Vanderplas- Support for scipy.spatial.cKDTree within neighbors queries has been removed, and the functionality replaced with the new
KDTreeclass.sklearn.neighbors.KernelDensityhas been added, which performs efficient kernel density estimation with a variety of kernels.sklearn.decomposition.KernelPCAnow always returns output withn_componentscomponents, unless the new parameterremove_zero_eigis set toTrue. This new behavior is consistent with the way kernel PCA was always documented; previously, the removal of components with zero eigenvalues was tacitly performed on all data.gcv_mode="auto"no longer tries to perform SVD on a densified sparse matrix insklearn.linear_model.RidgeCV.- Sparse matrix support in
sklearn.decomposition.RandomizedPCAis now deprecated in favor of the newTruncatedSVD.cross_validation.KFoldandcross_validation.StratifiedKFoldnow enforce n_folds >= 2 otherwise aValueErroris raised. By Olivier Grisel.datasets.load_files‘scharsetandcharset_errorsparameters were renamedencodinganddecode_errors.- Attribute
oob_score_insklearn.ensemble.GradientBoostingRegressorandsklearn.ensemble.GradientBoostingClassifieris deprecated and has been replaced byoob_improvement_.- Attributes in OrthogonalMatchingPursuit have been deprecated (copy_X, Gram, ...) and precompute_gram renamed precompute for consistency. See #2224.
sklearn.preprocessing.StandardScalernow converts integer input to float, and raises a warning. Previously it rounded for dense integer input.sklearn.multiclass.OneVsRestClassifiernow has adecision_functionmethod. This will return the distance of each sample from the decision boundary for each class, as long as the underlying estimators implement thedecision_functionmethod. By Kyle Kastner.- Better input validation, warning on unexpected shapes for y.

### People¶

List of contributors for release 0.14 by number of commits.

- 277 Gilles Louppe
- 245 Lars Buitinck
- 187 Andreas Mueller
- 124 Arnaud Joly
- 112 Jaques Grobler
- 109 Gael Varoquaux
- 107 Olivier Grisel
- 102 Noel Dawe
- 99 Kemal Eren
- 79 Joel Nothman
- 75 Jake VanderPlas
- 73 Nelle Varoquaux
- 71 Vlad Niculae
- 65 Peter Prettenhofer
- 64 Alexandre Gramfort
- 54 Mathieu Blondel
- 38 Nicolas Trésegnie
- 35 eustache
- 27 Denis Engemann
- 25 Yann N. Dauphin
- 19 Justin Vincent
- 17 Robert Layton
- 15 Doug Coleman
- 14 Michael Eickenberg
- 13 Robert Marchman
- 11 Fabian Pedregosa
- 11 Philippe Gervais
- 10 Jim Holmström
- 10 Tadej Janež
- 10 syhw
- 9 Mikhail Korobov
- 9 Steven De Gryze
- 8 sergeyf
- 7 Ben Root
- 7 Hrishikesh Huilgolkar
- 6 Kyle Kastner
- 6 Martin Luessi
- 6 Rob Speer
- 5 Federico Vaggi
- 5 Raul Garreta
- 5 Rob Zinkov
- 4 Ken Geis
- 3 A. Flaxman
- 3 Denton Cockburn
- 3 Dougal Sutherland
- 3 Ian Ozsvald
- 3 Johannes Schönberger
- 3 Robert McGibbon
- 3 Roman Sinayev
- 3 Szabo Roland
- 2 Diego Molla
- 2 Imran Haque
- 2 Jochen Wersdörfer
- 2 Sergey Karayev
- 2 Yannick Schwartz
- 2 jamestwebber
- 1 Abhijeet Kolhe
- 1 Alexander Fabisch
- 1 Bastiaan van den Berg
- 1 Benjamin Peterson
- 1 Daniel Velkov
- 1 Fazlul Shahriar
- 1 Felix Brockherde
- 1 Félix-Antoine Fortin
- 1 Harikrishnan S
- 1 Jack Hale
- 1 JakeMick
- 1 James McDermott
- 1 John Benediktsson
- 1 John Zwinck
- 1 Joshua Vredevoogd
- 1 Justin Pati
- 1 Kevin Hughes
- 1 Kyle Kelley
- 1 Matthias Ekman
- 1 Miroslav Shubernetskiy
- 1 Naoki Orii
- 1 Norbert Crombach
- 1 Rafael Cunha de Almeida
- 1 Rolando Espinoza La fuente
- 1 Seamus Abshere
- 1 Sergey Feldman
- 1 Sergio Medina
- 1 Stefano Lattarini
- 1 Steve Koch
- 1 Sturla Molden
- 1 Thomas Jarosch
- 1 Yaroslav Halchenko

## Version 0.13.1¶

The 0.13.1 release only fixes some bugs and does not add any new functionality.

### Changelog¶

- Fixed a testing error caused by the function
cross_validation.train_test_splitbeing interpreted as a test by Yaroslav Halchenko.- Fixed a bug in the reassignment of small clusters in the
cluster.MiniBatchKMeansby Gael Varoquaux.- Fixed default value of
gammaindecomposition.KernelPCAby Lars Buitinck.- Updated joblib to
0.7.0dby Gael Varoquaux.- Fixed scaling of the deviance in
ensemble.GradientBoostingClassifierby Peter Prettenhofer.- Better tie-breaking in
multiclass.OneVsOneClassifierby Andreas Müller.- Other small improvements to tests and documentation.

### People¶

- List of contributors for release 0.13.1 by number of commits.
- 16 Lars Buitinck
- 12 Andreas Müller
- 8 Gael Varoquaux
- 5 Robert Marchman
- 3 Peter Prettenhofer
- 2 Hrishikesh Huilgolkar
- 1 Bastiaan van den Berg
- 1 Diego Molla
- 1 Gilles Louppe
- 1 Mathieu Blondel
- 1 Nelle Varoquaux
- 1 Rafael Cunha de Almeida
- 1 Rolando Espinoza La fuente
- 1 Vlad Niculae
- 1 Yaroslav Halchenko

## Version 0.13¶

### New Estimator Classes¶

dummy.DummyClassifieranddummy.DummyRegressor, two data-independent predictors by Mathieu Blondel. Useful to sanity-check your estimators. SeeDummy estimatorsin the user guide. Multioutput support added by Arnaud Joly.decomposition.FactorAnalysis, a transformer implementing the classical factor analysis, by Christian Osendorfer and Alexandre Gramfort. SeeFactor Analysisin the user guide.feature_extraction.FeatureHasher, a transformer implementing the “hashing trick” for fast, low-memory feature extraction from string fields by Lars Buitinck andfeature_extraction.text.HashingVectorizerfor text documents by Olivier Grisel SeeFeature hashingandVectorizing a large text corpus with the hashing trickfor the documentation and sample usage.pipeline.FeatureUnion, a transformer that concatenates results of several other transformers by Andreas Müller. SeeFeatureUnion: composite feature spacesin the user guide.random_projection.GaussianRandomProjection,random_projection.SparseRandomProjectionand the functionrandom_projection.johnson_lindenstrauss_min_dim. The first two are transformers implementing Gaussian and sparse random projection matrix by Olivier Grisel and Arnaud Joly. SeeRandom Projectionin the user guide.kernel_approximation.Nystroem, a transformer for approximating arbitrary kernels by Andreas Müller. SeeNystroem Method for Kernel Approximationin the user guide.preprocessing.OneHotEncoder, a transformer that computes binary encodings of categorical features by Andreas Müller. SeeEncoding categorical featuresin the user guide.linear_model.PassiveAggressiveClassifierandlinear_model.PassiveAggressiveRegressor, predictors implementing an efficient stochastic optimization for linear models by Rob Zinkov and Mathieu Blondel. SeePassive Aggressive Algorithmsin the user guide.ensemble.RandomTreesEmbedding, a transformer for creating high-dimensional sparse representations using ensembles of totally random trees by Andreas Müller. SeeTotally Random Trees Embeddingin the user guide.manifold.SpectralEmbeddingand functionmanifold.spectral_embedding, implementing the “laplacian eigenmaps” transformation for non-linear dimensionality reduction by Wei Li. SeeSpectral Embeddingin the user guide.isotonic.IsotonicRegressionby Fabian Pedregosa, Alexandre Gramfort and Nelle Varoquaux,

### Changelog¶

metrics.zero_one_loss(formerlymetrics.zero_one) now has option for normalized output that reports the fraction of misclassifications, rather than the raw number of misclassifications. By Kyle Beauchamp.tree.DecisionTreeClassifierand all derived ensemble models now support sample weighting, by Noel Dawe and Gilles Louppe.- Speedup improvement when using bootstrap samples in forests of randomized trees, by Peter Prettenhofer and Gilles Louppe.
- Partial dependence plots for
Gradient Tree Boostinginensemble.partial_dependence.partial_dependenceby Peter Prettenhofer. SeePartial Dependence Plotsfor an example.- The table of contents on the website has now been made expandable by Jaques Grobler.
feature_selection.SelectPercentilenow breaks ties deterministically instead of returning all equally ranked features.feature_selection.SelectKBestandfeature_selection.SelectPercentileare more numerically stable since they use scores, rather than p-values, to rank results. This means that they might sometimes select different features than they did previously.- Ridge regression and ridge classification fitting with
sparse_cgsolver no longer has quadratic memory complexity, by Lars Buitinck and Fabian Pedregosa.- Ridge regression and ridge classification now support a new fast solver called
lsqr, by Mathieu Blondel.- Speed up of
metrics.precision_recall_curveby Conrad Lee.- Added support for reading/writing svmlight files with pairwise preference attribute (qid in svmlight file format) in
datasets.dump_svmlight_fileanddatasets.load_svmlight_fileby Fabian Pedregosa.- Faster and more robust
metrics.confusion_matrixandClustering performance evaluationby Wei Li.cross_validation.cross_val_scorenow works with precomputed kernels and affinity matrices, by Andreas Müller.- LARS algorithm made more numerically stable with heuristics to drop regressors too correlated as well as to stop the path when numerical noise becomes predominant, by Gael Varoquaux.
- Faster implementation of
metrics.precision_recall_curveby Conrad Lee.- New kernel
metrics.chi2_kernelby Andreas Müller, often used in computer vision applications.- Fix of longstanding bug in
naive_bayes.BernoulliNBfixed by Shaun Jackman.- Implemented
predict_probainmulticlass.OneVsRestClassifier, by Andrew Winterman.- Improve consistency in gradient boosting: estimators
ensemble.GradientBoostingRegressorandensemble.GradientBoostingClassifieruse the estimatortree.DecisionTreeRegressorinstead of thetree._tree.Treedata structure by Arnaud Joly.- Fixed a floating point exception in the
decision treesmodule, by Seberg.- Fix
metrics.roc_curvefails when y_true has only one class by Wei Li.- Add the
metrics.mean_absolute_errorfunction which computes the mean absolute error. Themetrics.mean_squared_error,metrics.mean_absolute_errorandmetrics.r2_scoremetrics support multioutput by Arnaud Joly.- Fixed
class_weightsupport insvm.LinearSVCandlinear_model.LogisticRegressionby Andreas Müller. The meaning ofclass_weightwas reversed as erroneously higher weight meant less positives of a given class in earlier releases.- Improve narrative documentation and consistency in
sklearn.metricsfor regression and classification metrics by Arnaud Joly.- Fixed a bug in
sklearn.svm.SVCwhen using csr-matrices with unsorted indices by Xinfan Meng and Andreas Müller.MiniBatchKMeans: Add random reassignment of cluster centers with little observations attached to them, by Gael Varoquaux.

### API changes summary¶

- Renamed all occurrences of
n_atomston_componentsfor consistency. This applies todecomposition.DictionaryLearning,decomposition.MiniBatchDictionaryLearning,decomposition.dict_learning,decomposition.dict_learning_online.- Renamed all occurrences of
max_iterstomax_iterfor consistency. This applies tosemi_supervised.LabelPropagationandsemi_supervised.label_propagation.LabelSpreading.- Renamed all occurrences of
learn_ratetolearning_ratefor consistency inensemble.BaseGradientBoostingandensemble.GradientBoostingRegressor.- The module
sklearn.linear_model.sparseis gone. Sparse matrix support was already integrated into the “regular” linear models.sklearn.metrics.mean_square_error, which incorrectly returned the accumulated error, was removed. Usemean_squared_errorinstead.- Passing
class_weightparameters tofitmethods is no longer supported. Pass them to estimator constructors instead.- GMMs no longer have
decodeandrvsmethods. Use thescore,predictorsamplemethods instead.- The
solverfit option in Ridge regression and classification is now deprecated and will be removed in v0.14. Use the constructor option instead.feature_extraction.text.DictVectorizernow returns sparse matrices in the CSR format, instead of COO.- Renamed
kincross_validation.KFoldandcross_validation.StratifiedKFoldton_folds, renamedn_bootstrapston_iterincross_validation.Bootstrap.- Renamed all occurrences of
n_iterationston_iterfor consistency. This applies tocross_validation.ShuffleSplit,cross_validation.StratifiedShuffleSplit,utils.randomized_range_finderandutils.randomized_svd.- Replaced
rhoinlinear_model.ElasticNetandlinear_model.SGDClassifierbyl1_ratio. Therhoparameter had different meanings;l1_ratiowas introduced to avoid confusion. It has the same meaning as previouslyrhoinlinear_model.ElasticNetand(1-rho)inlinear_model.SGDClassifier.linear_model.LassoLarsandlinear_model.Larsnow store a list of paths in the case of multiple targets, rather than an array of paths.- The attribute
gmmofhmm.GMMHMMwas renamed togmm_to adhere more strictly with the API.cluster.spectral_embeddingwas moved tomanifold.spectral_embedding.- Renamed
eig_tolinmanifold.spectral_embedding,cluster.SpectralClusteringtoeigen_tol, renamedmodetoeigen_solver.- Renamed
modeinmanifold.spectral_embeddingandcluster.SpectralClusteringtoeigen_solver.classes_andn_classes_attributes oftree.DecisionTreeClassifierand all derived ensemble models are now flat in case of single output problems and nested in case of multi-output problems.- The
estimators_attribute ofensemble.gradient_boosting.GradientBoostingRegressorandensemble.gradient_boosting.GradientBoostingClassifieris now an array of :class:’tree.DecisionTreeRegressor’.- Renamed
chunk_sizetobatch_sizeindecomposition.MiniBatchDictionaryLearninganddecomposition.MiniBatchSparsePCAfor consistency.svm.SVCandsvm.NuSVCnow provide aclasses_attribute and support arbitrary dtypes for labelsy. Also, the dtype returned bypredictnow reflects the dtype ofyduringfit(used to benp.float).- Changed default test_size in
cross_validation.train_test_splitto None, added possibility to infertest_sizefromtrain_sizeincross_validation.ShuffleSplitandcross_validation.StratifiedShuffleSplit.- Renamed function
sklearn.metrics.zero_onetosklearn.metrics.zero_one_loss. Be aware that the default behavior insklearn.metrics.zero_one_lossis different fromsklearn.metrics.zero_one:normalize=Falseis changed tonormalize=True.- Renamed function
metrics.zero_one_scoretometrics.accuracy_score.datasets.make_circlesnow has the same number of inner and outer points.- In the Naive Bayes classifiers, the
class_priorparameter was moved fromfitto__init__.

### People¶

List of contributors for release 0.13 by number of commits.

- 364 Andreas Müller
- 143 Arnaud Joly
- 137 Peter Prettenhofer
- 131 Gael Varoquaux
- 117 Mathieu Blondel
- 108 Lars Buitinck
- 106 Wei Li
- 101 Olivier Grisel
- 65 Vlad Niculae
- 54 Gilles Louppe
- 40 Jaques Grobler
- 38 Alexandre Gramfort
- 30 Rob Zinkov
- 19 Aymeric Masurelle
- 18 Andrew Winterman
- 17 Fabian Pedregosa
- 17 Nelle Varoquaux
- 16 Christian Osendorfer
- 14 Daniel Nouri
- 13 Virgile Fritsch
- 13 syhw
- 12 Satrajit Ghosh
- 10 Corey Lynch
- 10 Kyle Beauchamp
- 9 Brian Cheung
- 9 Immanuel Bayer
- 9 mr.Shu
- 8 Conrad Lee
- 8 James Bergstra
- 7 Tadej Janež
- 6 Brian Cajes
- 6 Jake Vanderplas
- 6 Michael
- 6 Noel Dawe
- 6 Tiago Nunes
- 6 cow
- 5 Anze
- 5 Shiqiao Du
- 4 Christian Jauvin
- 4 Jacques Kvam
- 4 Richard T. Guy
- 4 Robert Layton
- 3 Alexandre Abraham
- 3 Doug Coleman
- 3 Scott Dickerson
- 2 ApproximateIdentity
- 2 John Benediktsson
- 2 Mark Veronda
- 2 Matti Lyra
- 2 Mikhail Korobov
- 2 Xinfan Meng
- 1 Alejandro Weinstein
- 1 Alexandre Passos
- 1 Christoph Deil
- 1 Eugene Nizhibitsky
- 1 Kenneth C. Arnold
- 1 Luis Pedro Coelho
- 1 Miroslav Batchkarov
- 1 Pavel
- 1 Sebastian Berg
- 1 Shaun Jackman
- 1 Subhodeep Moitra
- 1 bob
- 1 dengemann
- 1 emanuele
- 1 x006

## Version 0.12.1¶

The 0.12.1 release is a bug-fix release with no additional features, but is instead a set of bug fixes

### Changelog¶

- Improved numerical stability in spectral embedding by Gael Varoquaux
- Doctest under windows 64bit by Gael Varoquaux
- Documentation fixes for elastic net by Andreas Müller and Alexandre Gramfort
- Proper behavior with fortran-ordered NumPy arrays by Gael Varoquaux
- Make GridSearchCV work with non-CSR sparse matrix by Lars Buitinck
- Fix parallel computing in MDS by Gael Varoquaux
- Fix Unicode support in count vectorizer by Andreas Müller
- Fix MinCovDet breaking with X.shape = (3, 1) by Virgile Fritsch
- Fix clone of SGD objects by Peter Prettenhofer
- Stabilize GMM by Virgile Fritsch

### People¶

## Version 0.12¶

### Changelog¶

- Various speed improvements of the
decision treesmodule, by Gilles Louppe.ensemble.GradientBoostingRegressorandensemble.GradientBoostingClassifiernow support feature subsampling via themax_featuresargument, by Peter Prettenhofer.- Added Huber and Quantile loss functions to
ensemble.GradientBoostingRegressor, by Peter Prettenhofer.Decision treesandforests of randomized treesnow support multi-output classification and regression problems, by Gilles Louppe.- Added
preprocessing.LabelEncoder, a simple utility class to normalize labels or transform non-numerical labels, by Mathieu Blondel.- Added the epsilon-insensitive loss and the ability to make probabilistic predictions with the modified huber loss in
Stochastic Gradient Descent, by Mathieu Blondel.- Added
Multi-dimensional Scaling (MDS), by Nelle Varoquaux.- SVMlight file format loader now detects compressed (gzip/bzip2) files and decompresses them on the fly, by Lars Buitinck.
- SVMlight file format serializer now preserves double precision floating point values, by Olivier Grisel.
- A common testing framework for all estimators was added, by Andreas Müller.
- Understandable error messages for estimators that do not accept sparse input by Gael Varoquaux
- Speedups in hierarchical clustering by Gael Varoquaux. In particular building the tree now supports early stopping. This is useful when the number of clusters is not small compared to the number of samples.
- Add MultiTaskLasso and MultiTaskElasticNet for joint feature selection, by Alexandre Gramfort.
- Added
metrics.auc_scoreandmetrics.average_precision_scoreconvenience functions by Andreas Müller.- Improved sparse matrix support in the
Feature selectionmodule by Andreas Müller.- New word boundaries-aware character n-gram analyzer for the
Text feature extractionmodule by @kernc.- Fixed bug in spectral clustering that led to single point clusters by Andreas Müller.
- In
feature_extraction.text.CountVectorizer, added an option to ignore infrequent words,min_dfby Andreas Müller.- Add support for multiple targets in some linear models (ElasticNet, Lasso and OrthogonalMatchingPursuit) by Vlad Niculae and Alexandre Gramfort.
- Fixes in
decomposition.ProbabilisticPCAscore function by Wei Li.- Fixed feature importance computation in
Gradient Tree Boosting.

### API changes summary¶

- The old
scikits.learnpackage has disappeared; all code should import fromsklearninstead, which was introduced in 0.9.- In
metrics.roc_curve, thethresholdsarray is now returned with it’s order reversed, in order to keep it consistent with the order of the returnedfprandtpr.- In
hmmobjects, likehmm.GaussianHMM,hmm.MultinomialHMM, etc., all parameters must be passed to the object when initialising it and not throughfit. Nowfitwill only accept the data as an input parameter.- For all SVM classes, a faulty behavior of
gammawas fixed. Previously, the default gamma value was only computed the first timefitwas called and then stored. It is now recalculated on every call tofit.- All
Baseclasses are now abstract meta classes so that they can not be instantiated.cluster.ward_treenow also returns the parent array. This is necessary for early-stopping in which case the tree is not completely built.- In
feature_extraction.text.CountVectorizerthe parametersmin_nandmax_nwere joined to the parametern_gram_rangeto enable grid-searching both at once.- In
feature_extraction.text.CountVectorizer, words that appear only in one document are now ignored by default. To reproduce the previous behavior, setmin_df=1.- Fixed API inconsistency:
linear_model.SGDClassifier.predict_probanow returns 2d array when fit on two classes.- Fixed API inconsistency:
discriminant_analysis.QuadraticDiscriminantAnalysis.decision_functionanddiscriminant_analysis.LinearDiscriminantAnalysis.decision_functionnow return 1d arrays when fit on two classes.- Grid of alphas used for fitting
linear_model.LassoCVandlinear_model.ElasticNetCVis now stored in the attributealphas_rather than overriding the init parameteralphas.- Linear models when alpha is estimated by cross-validation store the estimated value in the
alpha_attribute rather than justalphaorbest_alpha.ensemble.GradientBoostingClassifiernow supportsensemble.GradientBoostingClassifier.staged_predict_proba, andensemble.GradientBoostingClassifier.staged_predict.svm.sparse.SVCand other sparse SVM classes are now deprecated. The all classes in theSupport Vector Machinesmodule now automatically select the sparse or dense representation base on the input.- All clustering algorithms now interpret the array
Xgiven tofitas input data, in particularcluster.SpectralClusteringandcluster.AffinityPropagationwhich previously expected affinity matrices.- For clustering algorithms that take the desired number of clusters as a parameter, this parameter is now called
n_clusters.

### People¶

- 267 Andreas Müller
- 94 Gilles Louppe
- 89 Gael Varoquaux
- 79 Peter Prettenhofer
- 60 Mathieu Blondel
- 57 Alexandre Gramfort
- 52 Vlad Niculae
- 45 Lars Buitinck
- 44 Nelle Varoquaux
- 37 Jaques Grobler
- 30 Alexis Mignon
- 30 Immanuel Bayer
- 27 Olivier Grisel
- 16 Subhodeep Moitra
- 13 Yannick Schwartz
- 12 @kernc
- 11 Virgile Fritsch
- 9 Daniel Duckworth
- 9 Fabian Pedregosa
- 9 Robert Layton
- 8 John Benediktsson
- 7 Marko Burjek
- 5 Nicolas Pinto
- 4 Alexandre Abraham
- 4 Jake Vanderplas
- 3 Brian Holt
- 3 Edouard Duchesnay
- 3 Florian Hoenig
- 3 flyingimmidev
- 2 Francois Savard
- 2 Hannes Schulz
- 2 Peter Welinder
- 2 Yaroslav Halchenko
- 2 Wei Li
- 1 Alex Companioni
- 1 Brandyn A. White
- 1 Bussonnier Matthias
- 1 Charles-Pierre Astolfi
- 1 Dan O’Huiginn
- 1 David Cournapeau
- 1 Keith Goodman
- 1 Ludwig Schwardt
- 1 Olivier Hervieu
- 1 Sergio Medina
- 1 Shiqiao Du
- 1 Tim Sheerman-Chase
- 1 buguen

## Version 0.11¶

### Changelog¶

#### Highlights¶

- Gradient boosted regression trees (
Gradient Tree Boosting) for classification and regression by Peter Prettenhofer and Scott White .- Simple dict-based feature loader with support for categorical variables (
feature_extraction.DictVectorizer) by Lars Buitinck.- Added Matthews correlation coefficient (
metrics.matthews_corrcoef) and added macro and micro average options tometrics.precision_score,metrics.recall_scoreandmetrics.f1_scoreby Satrajit Ghosh.Out of Bag Estimatesof generalization error forEnsemble methodsby Andreas Müller.Randomized sparse models: Randomized sparse linear models for feature selection, by Alexandre Gramfort and Gael VaroquauxLabel Propagationfor semi-supervised learning, by Clay Woolam.Notethe semi-supervised API is still work in progress, and may change.- Added BIC/AIC model selection to classical
Gaussian mixture modelsand unified the API with the remainder of scikit-learn, by Bertrand Thirion- Added
sklearn.cross_validation.StratifiedShuffleSplit, which is asklearn.cross_validation.ShuffleSplitwith balanced splits, by Yannick Schwartz.sklearn.neighbors.NearestCentroidclassifier added, along with ashrink_thresholdparameter, which implementsshrunken centroid classification, by Robert Layton.

#### Other changes¶

- Merged dense and sparse implementations of
Stochastic Gradient Descentmodule and exposed utility extension types for sequential datasetsseq_datasetand weight vectorsweight_vectorby Peter Prettenhofer.- Added
partial_fit(support for online/minibatch learning) and warm_start to theStochastic Gradient Descentmodule by Mathieu Blondel.- Dense and sparse implementations of
Support Vector Machinesclasses andlinear_model.LogisticRegressionmerged by Lars Buitinck.- Regressors can now be used as base estimator in the
Multiclass and multilabel algorithmsmodule by Mathieu Blondel.- Added n_jobs option to
metrics.pairwise.pairwise_distancesandmetrics.pairwise.pairwise_kernelsfor parallel computation, by Mathieu Blondel.K-meanscan now be run in parallel, using then_jobsargument to eitherK-meansorKMeans, by Robert Layton.- Improved
Cross-validation: evaluating estimator performanceandTuning the hyper-parameters of an estimatordocumentation and introduced the newcross_validation.train_test_splithelper function by Olivier Griselsvm.SVCmemberscoef_andintercept_changed sign for consistency withdecision_function; forkernel==linear,coef_was fixed in the the one-vs-one case, by Andreas Müller.- Performance improvements to efficient leave-one-out cross-validated Ridge regression, esp. for the
n_samples > n_featurescase, inlinear_model.RidgeCV, by Reuben Fletcher-Costin.- Refactoring and simplification of the
Text feature extractionAPI and fixed a bug that caused possible negative IDF, by Olivier Grisel.- Beam pruning option in
_BaseHMMmodule has been removed since it is difficult to Cythonize. If you are interested in contributing a Cython version, you can use the python version in the git history as a reference.- Classes in
Nearest Neighborsnow support arbitrary Minkowski metric for nearest neighbors searches. The metric can be specified by argumentp.

### API changes summary¶

covariance.EllipticEnvelopis now deprecated - Please usecovariance.EllipticEnvelopeinstead.

NeighborsClassifierandNeighborsRegressorare gone in the moduleNearest Neighbors. Use the classesKNeighborsClassifier,RadiusNeighborsClassifier,KNeighborsRegressorand/orRadiusNeighborsRegressorinstead.Sparse classes in the

Stochastic Gradient Descentmodule are now deprecated.In

mixture.GMM,mixture.DPGMMandmixture.VBGMM, parameters must be passed to an object when initialising it and not throughfit. Nowfitwill only accept the data as an input parameter.methods

rvsanddecodeinGMMmodule are now deprecated.sampleandscoreorpredictshould be used instead.attribute

_scoresand_pvaluesin univariate feature selection objects are now deprecated.scores_orpvalues_should be used instead.In

LogisticRegression,LinearSVC,SVCandNuSVC, theclass_weightparameter is now an initialization parameter, not a parameter to fit. This makes grid searches over this parameter possible.LFW

datais now always shape(n_samples, n_features)to be consistent with the Olivetti faces dataset. Useimagesandpairsattribute to access the natural images shapes instead.In

svm.LinearSVC, the meaning of themulti_classparameter changed. Options now are'ovr'and'crammer_singer', with'ovr'being the default. This does not change the default behavior but hopefully is less confusing.Class

feature_selection.text.Vectorizeris deprecated and replaced byfeature_selection.text.TfidfVectorizer.The preprocessor / analyzer nested structure for text feature extraction has been removed. All those features are now directly passed as flat constructor arguments to

feature_selection.text.TfidfVectorizerandfeature_selection.text.CountVectorizer, in particular the following parameters are now used:

analyzercan be'word'or'char'to switch the default analysis scheme, or use a specific python callable (as previously).tokenizerandpreprocessorhave been introduced to make it still possible to customize those steps with the new API.inputexplicitly control how to interpret the sequence passed tofitandpredict: filenames, file objects or direct (byte or Unicode) strings.- charset decoding is explicit and strict by default.
- the
vocabulary, fitted or not is now stored in thevocabulary_attribute to be consistent with the project conventions.Class

feature_selection.text.TfidfVectorizernow derives directly fromfeature_selection.text.CountVectorizerto make grid search trivial.methods

rvsin_BaseHMMmodule are now deprecated.sampleshould be used instead.Beam pruning option in

_BaseHMMmodule is removed since it is difficult to be Cythonized. If you are interested, you can look in the history codes by git.The SVMlight format loader now supports files with both zero-based and one-based column indices, since both occur “in the wild”.

Arguments in class

ShuffleSplitare now consistent withStratifiedShuffleSplit. Argumentstest_fractionandtrain_fractionare deprecated and renamed totest_sizeandtrain_sizeand can accept bothfloatandint.Arguments in class

Bootstrapare now consistent withStratifiedShuffleSplit. Argumentsn_testandn_trainare deprecated and renamed totest_sizeandtrain_sizeand can accept bothfloatandint.Argument

padded to classes inNearest Neighborsto specify an arbitrary Minkowski metric for nearest neighbors searches.

### People¶

- 282 Andreas Müller
- 239 Peter Prettenhofer
- 198 Gael Varoquaux
- 129 Olivier Grisel
- 114 Mathieu Blondel
- 103 Clay Woolam
- 96 Lars Buitinck
- 88 Jaques Grobler
- 82 Alexandre Gramfort
- 50 Bertrand Thirion
- 42 Robert Layton
- 28 flyingimmidev
- 26 Jake Vanderplas
- 26 Shiqiao Du
- 21 Satrajit Ghosh
- 17 David Marek
- 17 Gilles Louppe
- 14 Vlad Niculae
- 11 Yannick Schwartz
- 10 Fabian Pedregosa
- 9 fcostin
- 7 Nick Wilson
- 5 Adrien Gaidon
- 5 Nicolas Pinto
- 4 David Warde-Farley
- 5 Nelle Varoquaux
- 5 Emmanuelle Gouillart
- 3 Joonas Sillanpää
- 3 Paolo Losi
- 2 Charles McCarthy
- 2 Roy Hyunjin Han
- 2 Scott White
- 2 ibayer
- 1 Brandyn White
- 1 Carlos Scheidegger
- 1 Claire Revillet
- 1 Conrad Lee
- 1 Edouard Duchesnay
- 1 Jan Hendrik Metzen
- 1 Meng Xinfan
- 1 Rob Zinkov
- 1 Shiqiao
- 1 Udi Weinsberg
- 1 Virgile Fritsch
- 1 Xinfan Meng
- 1 Yaroslav Halchenko
- 1 jansoe
- 1 Leon Palafox

## Version 0.10¶

### Changelog¶

- Python 2.5 compatibility was dropped; the minimum Python version needed to use scikit-learn is now 2.6.
Sparse inverse covarianceestimation using the graph Lasso, with associated cross-validated estimator, by Gael Varoquaux- New
Treemodule by Brian Holt, Peter Prettenhofer, Satrajit Ghosh and Gilles Louppe. The module comes with complete documentation and examples.- Fixed a bug in the RFE module by Gilles Louppe (issue #378).
- Fixed a memory leak in in
Support Vector Machinesmodule by Brian Holt (issue #367).- Faster tests by Fabian Pedregosa and others.
- Silhouette Coefficient cluster analysis evaluation metric added as
sklearn.metrics.silhouette_scoreby Robert Layton.- Fixed a bug in
K-meansin the handling of then_initparameter: the clustering algorithm used to be runn_inittimes but the last solution was retained instead of the best solution by Olivier Grisel.- Minor refactoring in
Stochastic Gradient Descentmodule; consolidated dense and sparse predict methods; Enhanced test time performance by converting model parameters to fortran-style arrays after fitting (only multi-class).- Adjusted Mutual Information metric added as
sklearn.metrics.adjusted_mutual_info_scoreby Robert Layton.- Models like SVC/SVR/LinearSVC/LogisticRegression from libsvm/liblinear now support scaling of C regularization parameter by the number of samples by Alexandre Gramfort.
- New
Ensemble Methodsmodule by Gilles Louppe and Brian Holt. The module comes with the random forest algorithm and the extra-trees method, along with documentation and examples.Novelty and Outlier Detection: outlier and novelty detection, by Virgile Fritsch.Kernel Approximation: a transform implementing kernel approximation for fast SGD on non-linear kernels by Andreas Müller.- Fixed a bug due to atom swapping in
Orthogonal Matching Pursuit (OMP)by Vlad Niculae.Sparse coding with a precomputed dictionaryby Vlad Niculae.Mini Batch K-Meansperformance improvements by Olivier Grisel.K-meanssupport for sparse matrices by Mathieu Blondel.- Improved documentation for developers and for the
sklearn.utilsmodule, by Jake Vanderplas.- Vectorized 20newsgroups dataset loader (
sklearn.datasets.fetch_20newsgroups_vectorized) by Mathieu Blondel.Multiclass and multilabel algorithmsby Lars Buitinck.- Utilities for fast computation of mean and variance for sparse matrices by Mathieu Blondel.
- Make
sklearn.preprocessing.scaleandsklearn.preprocessing.Scalerwork on sparse matrices by Olivier Grisel- Feature importances using decision trees and/or forest of trees, by Gilles Louppe.
- Parallel implementation of forests of randomized trees by Gilles Louppe.
sklearn.cross_validation.ShuffleSplitcan subsample the train sets as well as the test sets by Olivier Grisel.- Errors in the build of the documentation fixed by Andreas Müller.

### API changes summary¶

Here are the code migration instructions when upgrading from scikit-learn version 0.9:

Some estimators that may overwrite their inputs to save memory previously had

overwrite_parameters; these have been replaced withcopy_parameters with exactly the opposite meaning.This particularly affects some of the estimators in

linear_model. The default behavior is still to copy everything passed in.The SVMlight dataset loader

sklearn.datasets.load_svmlight_fileno longer supports loading two files at once; useload_svmlight_filesinstead. Also, the (unused)buffer_mbparameter is gone.Sparse estimators in the

Stochastic Gradient Descentmodule use dense parameter vectorcoef_instead ofsparse_coef_. This significantly improves test time performance.The

Covariance estimationmodule now has a robust estimator of covariance, the Minimum Covariance Determinant estimator.Cluster evaluation metrics in

metrics.clusterhave been refactored but the changes are backwards compatible. They have been moved to themetrics.cluster.supervised, along withmetrics.cluster.unsupervisedwhich contains the Silhouette Coefficient.The

permutation_test_scorefunction now behaves the same way ascross_val_score(i.e. uses the mean score across the folds.)Cross Validation generators now use integer indices (

indices=True) by default instead of boolean masks. This make it more intuitive to use with sparse matrix data.The functions used for sparse coding,

sparse_encodeandsparse_encode_parallelhave been combined intosklearn.decomposition.sparse_encode, and the shapes of the arrays have been transposed for consistency with the matrix factorization setting, as opposed to the regression setting.Fixed an off-by-one error in the SVMlight/LibSVM file format handling; files generated using

sklearn.datasets.dump_svmlight_fileshould be re-generated. (They should continue to work, but accidentally had one extra column of zeros prepended.)

BaseDictionaryLearningclass replaced bySparseCodingMixin.

sklearn.utils.extmath.fast_svdhas been renamedsklearn.utils.extmath.randomized_svdand the default oversampling is now fixed to 10 additional random vectors instead of doubling the number of components to extract. The new behavior follows the reference paper.

### People¶

The following people contributed to scikit-learn since last release:

- 246 Andreas Müller
- 242 Olivier Grisel
- 220 Gilles Louppe
- 183 Brian Holt
- 166 Gael Varoquaux
- 144 Lars Buitinck
- 73 Vlad Niculae
- 65 Peter Prettenhofer
- 64 Fabian Pedregosa
- 60 Robert Layton
- 55 Mathieu Blondel
- 52 Jake Vanderplas
- 44 Noel Dawe
- 38 Alexandre Gramfort
- 24 Virgile Fritsch
- 23 Satrajit Ghosh
- 3 Jan Hendrik Metzen
- 3 Kenneth C. Arnold
- 3 Shiqiao Du
- 3 Tim Sheerman-Chase
- 3 Yaroslav Halchenko
- 2 Bala Subrahmanyam Varanasi
- 2 DraXus
- 2 Michael Eickenberg
- 1 Bogdan Trach
- 1 Félix-Antoine Fortin
- 1 Juan Manuel Caicedo Carvajal
- 1 Nelle Varoquaux
- 1 Nicolas Pinto
- 1 Tiziano Zito
- 1 Xinfan Meng

## Version 0.9¶

scikit-learn 0.9 was released on September 2011, three months after the 0.8
release and includes the new modules *Manifold learning*, *The Dirichlet Process*
as well as several new algorithms and documentation improvements.

This release also includes the dictionary-learning work developed by Vlad Niculae as part of the Google Summer of Code program.

### Changelog¶

- New
Manifold learningmodule by Jake Vanderplas and Fabian Pedregosa.- New
Dirichlet ProcessGaussian Mixture Model by Alexandre PassosNearest Neighborsmodule refactoring by Jake Vanderplas : general refactoring, support for sparse matrices in input, speed and documentation improvements. See the next section for a full list of API changes.- Improvements on the
Feature selectionmodule by Gilles Louppe : refactoring of the RFE classes, documentation rewrite, increased efficiency and minor API changes.Sparse principal components analysis (SparsePCA and MiniBatchSparsePCA)by Vlad Niculae, Gael Varoquaux and Alexandre Gramfort- Printing an estimator now behaves independently of architectures and Python version thanks to Jean Kossaifi.
Loader for libsvm/svmlight formatby Mathieu Blondel and Lars Buitinck- Documentation improvements: thumbnails in
example galleryby Fabian Pedregosa.- Important bugfixes in
Support Vector Machinesmodule (segfaults, bad performance) by Fabian Pedregosa.- Added
Multinomial Naive BayesandBernoulli Naive Bayesby Lars Buitinck- Text feature extraction optimizations by Lars Buitinck
- Chi-Square feature selection (
feature_selection.univariate_selection.chi2) by Lars Buitinck.Sample generatorsmodule refactoring by Gilles LouppeMulticlass and multilabel algorithmsby Mathieu Blondel- Ball tree rewrite by Jake Vanderplas
- Implementation of
DBSCANalgorithm by Robert Layton- Kmeans predict and transform by Robert Layton
- Preprocessing module refactoring by Olivier Grisel
- Faster mean shift by Conrad Lee
- New
Bootstrap,Random permutations cross-validation a.k.a. Shuffle & Splitand various other improvements in cross validation schemes by Olivier Grisel and Gael Varoquaux- Adjusted Rand index and V-Measure clustering evaluation metrics by Olivier Grisel
- Added
Orthogonal Matching Pursuitby Vlad Niculae- Added 2D-patch extractor utilities in the
Feature extractionmodule by Vlad Niculae- Implementation of
linear_model.LassoLarsCV(cross-validated Lasso solver using the Lars algorithm) andlinear_model.LassoLarsIC(BIC/AIC model selection in Lars) by Gael Varoquaux and Alexandre Gramfort- Scalability improvements to
metrics.roc_curveby Olivier Hervieu- Distance helper functions
metrics.pairwise.pairwise_distancesandmetrics.pairwise.pairwise_kernelsby Robert LaytonMini-Batch K-Meansby Nelle Varoquaux and Peter Prettenhofer.Downloading datasets from the mldata.org repositoryutilities by Pietro Berkes.The Olivetti faces datasetby David Warde-Farley.

### API changes summary¶

Here are the code migration instructions when upgrading from scikit-learn version 0.8:

The

scikits.learnpackage was renamedsklearn. There is still ascikits.learnpackage alias for backward compatibility.Third-party projects with a dependency on scikit-learn 0.9+ should upgrade their codebase. For instance under Linux / MacOSX just run (make a backup first!):

find -name "*.py" | xargs sed -i 's/\bscikits.learn\b/sklearn/g'Estimators no longer accept model parameters as

fitarguments: instead all parameters must be only be passed as constructor arguments or using the now publicset_paramsmethod inherited frombase.BaseEstimator.Some estimators can still accept keyword arguments on the

fitbut this is restricted to data-dependent values (e.g. a Gram matrix or an affinity matrix that are precomputed from theXdata matrix.The

cross_valpackage has been renamed tocross_validationalthough there is also across_valpackage alias in place for backward compatibility.Third-party projects with a dependency on scikit-learn 0.9+ should upgrade their codebase. For instance under Linux / MacOSX just run (make a backup first!):

find -name "*.py" | xargs sed -i 's/\bcross_val\b/cross_validation/g'The

score_funcargument of thesklearn.cross_validation.cross_val_scorefunction is now expected to accepty_testandy_predictedas only arguments for classification and regression tasks orX_testfor unsupervised estimators.

gammaparameter for support vector machine algorithms is set to1 / n_featuresby default, instead of1 / n_samples.The

sklearn.hmmhas been marked as orphaned: it will be removed from scikit-learn in version 0.11 unless someone steps up to contribute documentation, examples and fix lurking numerical stability issues.

sklearn.neighborshas been made into a submodule. The two previously available estimators,NeighborsClassifierandNeighborsRegressorhave been marked as deprecated. Their functionality has been divided among five new classes:NearestNeighborsfor unsupervised neighbors searches,KNeighborsClassifier&RadiusNeighborsClassifierfor supervised classification problems, andKNeighborsRegressor&RadiusNeighborsRegressorfor supervised regression problems.

sklearn.ball_tree.BallTreehas been moved tosklearn.neighbors.BallTree. Using the former will generate a warning.

sklearn.linear_model.LARS()and related classes (LassoLARS, LassoLARSCV, etc.) have been renamed tosklearn.linear_model.Lars().All distance metrics and kernels in

sklearn.metrics.pairwisenow have a Y parameter, which by default is None. If not given, the result is the distance (or kernel similarity) between each sample in Y. If given, the result is the pairwise distance (or kernel similarity) between samples in X to Y.

sklearn.metrics.pairwise.l1_distanceis now calledmanhattan_distance, and by default returns the pairwise distance. For the component wise distance, set the parametersum_over_featurestoFalse.

Backward compatibility package aliases and other deprecated classes and functions will be removed in version 0.11.

### People¶

38 people contributed to this release.

- 387 Vlad Niculae
- 320 Olivier Grisel
- 192 Lars Buitinck
- 179 Gael Varoquaux
- 168 Fabian Pedregosa (INRIA, Parietal Team)
- 127 Jake Vanderplas
- 120 Mathieu Blondel
- 85 Alexandre Passos
- 67 Alexandre Gramfort
- 57 Peter Prettenhofer
- 56 Gilles Louppe
- 42 Robert Layton
- 38 Nelle Varoquaux
- 32 Jean Kossaifi
- 30 Conrad Lee
- 22 Pietro Berkes
- 18 andy
- 17 David Warde-Farley
- 12 Brian Holt
- 11 Robert
- 8 Amit Aides
- 8 Virgile Fritsch
- 7 Yaroslav Halchenko
- 6 Salvatore Masecchia
- 5 Paolo Losi
- 4 Vincent Schut
- 3 Alexis Metaireau
- 3 Bryan Silverthorn
- 3 Andreas Müller
- 2 Minwoo Jake Lee
- 1 Emmanuelle Gouillart
- 1 Keith Goodman
- 1 Lucas Wiman
- 1 Nicolas Pinto
- 1 Thouis (Ray) Jones
- 1 Tim Sheerman-Chase

## Version 0.8¶

scikit-learn 0.8 was released on May 2011, one month after the first
“international” scikit-learn coding sprint and is
marked by the inclusion of important modules: *Hierarchical clustering*,
*Cross decomposition*, *Non-negative matrix factorization (NMF or NNMF)*, initial support for Python 3 and by important
enhancements and bug fixes.

### Changelog¶

Several new modules where introduced during this release:

- New
Hierarchical clusteringmodule by Vincent Michel, Bertrand Thirion, Alexandre Gramfort and Gael Varoquaux.Kernel PCAimplementation by Mathieu BlondelThe Labeled Faces in the Wild face recognition datasetby Olivier Grisel.- New
Cross decompositionmodule by Edouard Duchesnay.Non-negative matrix factorization (NMF or NNMF)module Vlad Niculae- Implementation of the
Oracle Approximating Shrinkagealgorithm by Virgile Fritsch in theCovariance estimationmodule.

Some other modules benefited from significant improvements or cleanups.

- Initial support for Python 3: builds and imports cleanly, some modules are usable while others have failing tests by Fabian Pedregosa.
decomposition.PCAis now usable from the Pipeline object by Olivier Grisel.- Guide
How to optimize for speedby Olivier Grisel.- Fixes for memory leaks in libsvm bindings, 64-bit safer BallTree by Lars Buitinck.
- bug and style fixing in
K-meansalgorithm by Jan Schlüter.- Add attribute converged to Gaussian Mixture Models by Vincent Schut.
- Implemented
transform,predict_log_probaindiscriminant_analysis.LinearDiscriminantAnalysisBy Mathieu Blondel.- Refactoring in the
Support Vector Machinesmodule and bug fixes by Fabian Pedregosa, Gael Varoquaux and Amit Aides.- Refactored SGD module (removed code duplication, better variable naming), added interface for sample weight by Peter Prettenhofer.
- Wrapped BallTree with Cython by Thouis (Ray) Jones.
- Added function
svm.l1_min_cby Paolo Losi.- Typos, doc style, etc. by Yaroslav Halchenko, Gael Varoquaux, Olivier Grisel, Yann Malet, Nicolas Pinto, Lars Buitinck and Fabian Pedregosa.

### People¶

People that made this release possible preceded by number of commits:

- 159 Olivier Grisel
- 96 Gael Varoquaux
- 96 Vlad Niculae
- 94 Fabian Pedregosa
- 36 Alexandre Gramfort
- 32 Paolo Losi
- 31 Edouard Duchesnay
- 30 Mathieu Blondel
- 25 Peter Prettenhofer
- 22 Nicolas Pinto
- 11 Virgile Fritsch
- 7 Lars Buitinck
- 6 Vincent Michel
- 5 Bertrand Thirion
- 4 Thouis (Ray) Jones
- 4 Vincent Schut
- 3 Jan Schlüter
- 2 Julien Miotte
- 2 Matthieu Perrot
- 2 Yann Malet
- 2 Yaroslav Halchenko
- 1 Amit Aides
- 1 Andreas Müller
- 1 Feth Arezki
- 1 Meng Xinfan

## Version 0.7¶

scikit-learn 0.7 was released in March 2011, roughly three months after the 0.6 release. This release is marked by the speed improvements in existing algorithms like k-Nearest Neighbors and K-Means algorithm and by the inclusion of an efficient algorithm for computing the Ridge Generalized Cross Validation solution. Unlike the preceding release, no new modules where added to this release.

### Changelog¶

- Performance improvements for Gaussian Mixture Model sampling [Jan Schlüter].
- Implementation of efficient leave-one-out cross-validated Ridge in
linear_model.RidgeCV[Mathieu Blondel]- Better handling of collinearity and early stopping in
linear_model.lars_path[Alexandre Gramfort and Fabian Pedregosa].- Fixes for liblinear ordering of labels and sign of coefficients [Dan Yamins, Paolo Losi, Mathieu Blondel and Fabian Pedregosa].
- Performance improvements for Nearest Neighbors algorithm in high-dimensional spaces [Fabian Pedregosa].
- Performance improvements for
cluster.KMeans[Gael Varoquaux and James Bergstra].- Sanity checks for SVM-based classes [Mathieu Blondel].
- Refactoring of
neighbors.NeighborsClassifierandneighbors.kneighbors_graph: added different algorithms for the k-Nearest Neighbor Search and implemented a more stable algorithm for finding barycenter weights. Also added some developer documentation for this module, see notes_neighbors for more information [Fabian Pedregosa].- Documentation improvements: Added
pca.RandomizedPCAandlinear_model.LogisticRegressionto the class reference. Also added references of matrices used for clustering and other fixes [Gael Varoquaux, Fabian Pedregosa, Mathieu Blondel, Olivier Grisel, Virgile Fritsch , Emmanuelle Gouillart]- Binded decision_function in classes that make use of liblinear, dense and sparse variants, like
svm.LinearSVCorlinear_model.LogisticRegression[Fabian Pedregosa].- Performance and API improvements to
metrics.euclidean_distancesand topca.RandomizedPCA[James Bergstra].- Fix compilation issues under NetBSD [Kamel Ibn Hassen Derouiche]
- Allow input sequences of different lengths in
hmm.GaussianHMM[Ron Weiss].- Fix bug in affinity propagation caused by incorrect indexing [Xinfan Meng]

### People¶

People that made this release possible preceded by number of commits:

- 85 Fabian Pedregosa
- 67 Mathieu Blondel
- 20 Alexandre Gramfort
- 19 James Bergstra
- 14 Dan Yamins
- 13 Olivier Grisel
- 12 Gael Varoquaux
- 4 Edouard Duchesnay
- 4 Ron Weiss
- 2 Satrajit Ghosh
- 2 Vincent Dubourg
- 1 Emmanuelle Gouillart
- 1 Kamel Ibn Hassen Derouiche
- 1 Paolo Losi
- 1 VirgileFritsch
- 1 Yaroslav Halchenko
- 1 Xinfan Meng

## Version 0.6¶

scikit-learn 0.6 was released on December 2010. It is marked by the inclusion of several new modules and a general renaming of old ones. It is also marked by the inclusion of new example, including applications to real-world datasets.

### Changelog¶

- New stochastic gradient descent module by Peter Prettenhofer. The module comes with complete documentation and examples.
- Improved svm module: memory consumption has been reduced by 50%, heuristic to automatically set class weights, possibility to assign weights to samples (see
SVM: Weighted samplesfor an example).- New
Gaussian Processesmodule by Vincent Dubourg. This module also has great documentation and some very neat examples. See example_gaussian_process_plot_gp_regression.py or example_gaussian_process_plot_gp_probabilistic_classification_after_regression.py for a taste of what can be done.- It is now possible to use liblinear’s Multi-class SVC (option multi_class in
svm.LinearSVC)- New features and performance improvements of text feature extraction.
- Improved sparse matrix support, both in main classes (
grid_search.GridSearchCV) as in modules sklearn.svm.sparse and sklearn.linear_model.sparse.- Lots of cool new examples and a new section that uses real-world datasets was created. These include:
Faces recognition example using eigenfaces and SVMs,Species distribution modeling,Libsvm GUI,Wikipedia principal eigenvectorand others.- Faster
Least Angle Regressionalgorithm. It is now 2x faster than the R version on worst case and up to 10x times faster on some cases.- Faster coordinate descent algorithm. In particular, the full path version of lasso (
linear_model.lasso_path) is more than 200x times faster than before.- It is now possible to get probability estimates from a
linear_model.LogisticRegressionmodel.- module renaming: the glm module has been renamed to linear_model, the gmm module has been included into the more general mixture model and the sgd module has been included in linear_model.
- Lots of bug fixes and documentation improvements.

### People¶

People that made this release possible preceded by number of commits:

- 207 Olivier Grisel
- 167 Fabian Pedregosa
- 97 Peter Prettenhofer
- 68 Alexandre Gramfort
- 59 Mathieu Blondel
- 55 Gael Varoquaux
- 33 Vincent Dubourg
- 21 Ron Weiss
- 9 Bertrand Thirion
- 3 Alexandre Passos
- 3 Anne-Laure Fouque
- 2 Ronan Amicel
- 1 Christian Osendorfer

## Version 0.5¶

### Changelog¶

### New classes¶

- Support for sparse matrices in some classifiers of modules
svmandlinear_model(seesvm.sparse.SVC,svm.sparse.SVR,svm.sparse.LinearSVC,linear_model.sparse.Lasso,linear_model.sparse.ElasticNet)- New
pipeline.Pipelineobject to compose different estimators.- Recursive Feature Elimination routines in module
Feature selection.- Addition of various classes capable of cross validation in the linear_model module (
linear_model.LassoCV,linear_model.ElasticNetCV, etc.).- New, more efficient LARS algorithm implementation. The Lasso variant of the algorithm is also implemented. See
linear_model.lars_path,linear_model.Larsandlinear_model.LassoLars.- New Hidden Markov Models module (see classes
hmm.GaussianHMM,hmm.MultinomialHMM,hmm.GMMHMM)- New module feature_extraction (see
class reference)- New FastICA algorithm in module sklearn.fastica

### Documentation¶

- Improved documentation for many modules, now separating narrative documentation from the class reference. As an example, see documentation for the SVM module and the complete class reference.

### Fixes¶

- API changes: adhere variable names to PEP-8, give more meaningful names.
- Fixes for svm module to run on a shared memory context (multiprocessing).
- It is again possible to generate latex (and thus PDF) from the sphinx docs.

### Examples¶

- new examples using some of the mlcomp datasets:
example_mlcomp_sparse_document_classification.py(since removed) andClassification of text documents using sparse features- Many more examples. See here the full list of examples.

### External dependencies¶

- Joblib is now a dependency of this package, although it is shipped with (sklearn.externals.joblib).

### Removed modules¶

- Module ann (Artificial Neural Networks) has been removed from the distribution. Users wanting this sort of algorithms should take a look into pybrain.

### Misc¶

- New sphinx theme for the web page.

### Authors¶

The following is a list of authors for this release, preceded by number of commits:

- 262 Fabian Pedregosa
- 240 Gael Varoquaux
- 149 Alexandre Gramfort
- 116 Olivier Grisel
- 40 Vincent Michel
- 38 Ron Weiss
- 23 Matthieu Perrot
- 10 Bertrand Thirion
- 7 Yaroslav Halchenko
- 9 VirgileFritsch
- 6 Edouard Duchesnay
- 4 Mathieu Blondel
- 1 Ariel Rokem
- 1 Matthieu Brucher

## Version 0.4¶

### Changelog¶

Major changes in this release include:

- Coordinate Descent algorithm (Lasso, ElasticNet) refactoring & speed improvements (roughly 100x times faster).
- Coordinate Descent Refactoring (and bug fixing) for consistency with R’s package GLMNET.
- New metrics module.
- New GMM module contributed by Ron Weiss.
- Implementation of the LARS algorithm (without Lasso variant for now).
- feature_selection module redesign.
- Migration to GIT as version control system.
- Removal of obsolete attrselect module.
- Rename of private compiled extensions (added underscore).
- Removal of legacy unmaintained code.
- Documentation improvements (both docstring and rst).
- Improvement of the build system to (optionally) link with MKL. Also, provide a lite BLAS implementation in case no system-wide BLAS is found.
- Lots of new examples.
- Many, many bug fixes ...

### Authors¶

The committer list for this release is the following (preceded by number of commits):

- 143 Fabian Pedregosa
- 35 Alexandre Gramfort
- 34 Olivier Grisel
- 11 Gael Varoquaux
- 5 Yaroslav Halchenko
- 2 Vincent Michel
- 1 Chris Filo Gorgolewski

## Earlier versions¶

Earlier versions included contributions by Fred Mailhot, David Cooke, David Huard, Dave Morrill, Ed Schofield, Travis Oliphant, Pearu Peterson.