Older Versions
scikit-learn Docs · نسخهٔ 5e9184e53cd3 · BSD-3-Clause
==============
Older Versions
Version 0.12.1
October 8, 2012
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 <VirgileFritsch>
- Fix clone of SGD objects by
Peter Prettenhofer_
- Stabilize GMM by
Virgile Fritsch <VirgileFritsch>
People
- 14
Peter Prettenhofer_ - 12
Gael Varoquaux_ - 10
Andreas Müller_ - 5
Lars Buitinck_ - 3
Virgile Fritsch <VirgileFritsch> - 1
Alexandre Gramfort_ - 1
Gilles Louppe_ - 1
Mathieu Blondel_
Version 0.12
September 4, 2012
Changelog
- Various speed improvements of the
decision trees <tree>module, by
Gilles Louppe_.
~ensemble.GradientBoostingRegressorand
~ensemble.GradientBoostingClassifier now support feature subsampling via the `max_features argument, by Peter Prettenhofer`_.
- Added Huber and Quantile loss functions to
~ensemble.GradientBoostingRegressor, by Peter Prettenhofer_.
Decision trees <tree>andforests of randomized trees <forest>
now 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 sgd, by Mathieu Blondel_.
- Added
multidimensional_scaling, 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_scoreand
metrics.average_precision_score convenience functions by Andreas Müller_.
- Improved sparse matrix support in the
feature_selection
module by Andreas Müller_.
- New word boundaries-aware character n-gram analyzer for the
text_feature_extraction module by @kernc <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_df by 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_boosting.
API changes summary
- The old `
scikits.learn` package has disappeared; all code should import
from `sklearn` instead, which was introduced in 0.9.
- In
metrics.roc_curve, the `thresholds` array is now returned
with its order reversed, in order to keep it consistent with the order of the returned `fpr and tpr`.
- In
hmmobjects, likehmm.GaussianHMM,
hmm.MultinomialHMM, etc., all parameters must be passed to the object when initialising it and not through `fit. Now fit` will only accept the data as an input parameter.
- For all SVM classes, a faulty behavior of `
gamma` was fixed. Previously,
the default gamma value was only computed the first time `fit was called and then stored. It is now recalculated on every call to fit`.
- All `
Base` classes 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 parameters
`min_n and max_n were joined to the parameter n_gram_range` to 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, set `min_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_function
and discriminant_analysis.LinearDiscriminantAnalysis.decision_function now return 1d arrays when fit on two classes.
- Grid of alphas used for fitting
~linear_model.LassoCVand
~linear_model.ElasticNetCV is now stored in the attribute `alphas_ rather than overriding the init parameter alphas`.
- Linear models when alpha is estimated by cross-validation store
the estimated value in the `alpha_ attribute rather than just alpha or best_alpha`.
~ensemble.GradientBoostingClassifiernow supports
~ensemble.GradientBoostingClassifier.staged_predict_proba, and ~ensemble.GradientBoostingClassifier.staged_predict.
svm.sparse.SVCand other sparse SVM classes are now deprecated.
All classes in the svm module now automatically select the sparse or dense representation based on the input.
- All clustering algorithms now interpret the array `
Xgiven tofit` as
input data, in particular ~cluster.SpectralClustering and ~cluster.AffinityPropagation which 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 <kernc> - 11
Virgile Fritsch <VirgileFritsch> - 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
May 7, 2012
Changelog
Highlights .............
- Gradient boosted regression trees (
gradient_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 to ~metrics.precision_score, metrics.recall_score and ~metrics.f1_score by Satrajit Ghosh_.
out_of_bagof generalization error forensemble
by Andreas Müller_.
- Randomized sparse linear models for feature
selection, by Alexandre Gramfort_ and Gael Varoquaux_
label_propagationfor semi-supervised learning, by Clay
Woolam. Note the semi-supervised API is still work in progress, and may change.
- Added BIC/AIC model selection to classical
gmmand unified
the API with the remainder of scikit-learn, by Bertrand Thirion_
- Added
sklearn.cross_validation.StratifiedShuffleSplit, which is
a sklearn.cross_validation.ShuffleSplit with balanced splits, by Yannick Schwartz.
~sklearn.neighbors.NearestCentroidclassifier added, along with a
`shrink_threshold parameter, which implements **shrunken centroid classification**, by Robert Layton`_.
Other changes ..............
- Merged dense and sparse implementations of
sgdmodule and
exposed utility extension types for sequential datasets `seq_dataset and weight vectors weight_vector by Peter Prettenhofer`_.
- Added `
partial_fit` (support for online/minibatch learning) and
warm_start to the sgd module by Mathieu Blondel_.
- Dense and sparse implementations of
svmclasses and
~linear_model.LogisticRegression merged by Lars Buitinck_.
- Regressors can now be used as base estimator in the
multiclass
module by Mathieu Blondel_.
- Added n_jobs option to
metrics.pairwise_distances
and metrics.pairwise.pairwise_kernels for parallel computation, by Mathieu Blondel_.
k_meanscan now be run in parallel, using the `n_jobs` argument
to either k_means or cluster.KMeans, by Robert Layton_.
- Improved
cross_validationandgrid_searchdocumentation
and introduced the new cross_validation.train_test_split helper function by Olivier Grisel_
~svm.SVCmembers `coef_andintercept_` changed sign for
consistency with `decision_function; for kernel==linear, coef_ was fixed in 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_features case, in ~linear_model.RidgeCV`, by Reuben Fletcher-Costin.
- Refactoring and simplification of the
text_feature_extraction
API 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
neighborsnow support arbitrary Minkowski metric for
nearest neighbors searches. The metric can be specified by argument `p`.
API changes summary
covariance.EllipticEnvelopis now deprecated.
Please use ~covariance.EllipticEnvelope instead.
- `
NeighborsClassifierandNeighborsRegressor` are gone in the module
neighbors. Use the classes ~neighbors.KNeighborsClassifier, ~neighbors.RadiusNeighborsClassifier, ~neighbors.KNeighborsRegressor and/or ~neighbors.RadiusNeighborsRegressor instead.
- Sparse classes in the
sgdmodule are now deprecated.
- In
mixture.GMM,mixture.DPGMMandmixture.VBGMM,
parameters must be passed to an object when initialising it and not through `fit. Now fit` will only accept the data as an input parameter.
- methods `
rvsanddecodeinGMM` module are now deprecated.
`sample and score or predict` should be used instead.
- attribute `
_scoresand_pvalues` in univariate feature selection
objects are now deprecated. `scores_ or pvalues_` should be used instead.
- In
~linear_model.LogisticRegression,~svm.LinearSVC,
~svm.SVC and ~svm.NuSVC, the `class_weight` parameter 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. Use `images and pairs` attribute to access the natural images shapes instead.
- In
~svm.LinearSVC, the meaning of the `multi_class` parameter
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 by feature_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.TfidfVectorizer and feature_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).
- `
tokenizerandpreprocessor` have been introduced to make it
still possible to customize those steps with the new API.
- `
input` explicitly control how to interpret the sequence passed to
`fit and predict`: 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 the
`vocabulary_` attribute to be consistent with the project conventions.
- Class
feature_selection.text.TfidfVectorizernow derives directly
from feature_selection.text.CountVectorizer to make grid search trivial.
- methods `
rvsin_BaseHMM` module are now deprecated.
`sample` should 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
~model_selection.ShuffleSplitare now consistent with
~model_selection.StratifiedShuffleSplit. Arguments `test_fraction and train_fraction are deprecated and renamed to test_size and train_size and can accept both float and int`.
- Arguments in class
Bootstrapare now consistent with
~model_selection.StratifiedShuffleSplit. Arguments `n_test and n_train are deprecated and renamed to test_size and train_size and can accept both float and int`.
- Argument `
padded to classes inneighbors` to 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
January 11, 2012
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
Tree <tree>module byBrian 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
svmmodule byBrian Holt_ (issue #367).
- Faster tests by
Fabian Pedregosa_ and others.
- Silhouette Coefficient cluster analysis evaluation metric added as
~sklearn.metrics.silhouette_score by Robert Layton.
- Fixed a bug in
k_meansin the handling of the `n_init` parameter:
the clustering algorithm used to be run `n_init times but the last solution was retained instead of the best solution by Olivier Grisel`_.
- Minor refactoring in
sgdmodule; 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_score by 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 Methods <ensemble>module byGilles Louppe_ and
Brian Holt_. The module comes with the random forest algorithm and the extra-trees method, along with documentation and examples.
outlier_detection: outlier and novelty detection, by
Virgile Fritsch <VirgileFritsch>.
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
OMPbyVlad Niculae_.
SparseCoderbyVlad Niculae_.
mini_batch_kmeansperformance improvements byOlivier Grisel_.
k_meanssupport for sparse matrices byMathieu Blondel_.
- Improved documentation for developers and for the
sklearn.utils
module, by Jake Vanderplas_.
- Vectorized 20newsgroups dataset loader
(~sklearn.datasets.fetch_20newsgroups_vectorized) by Mathieu Blondel_.
multiclassbyLars Buitinck_.
- Utilities for fast computation of mean and variance for sparse matrices
by Mathieu Blondel_.
- Make
~sklearn.preprocessing.scaleand
sklearn.preprocessing.Scaler work 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 with copy_` parameters with exactly the opposite meaning.
This particularly affects some of the estimators in ~sklearn.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; use `load_svmlight_files instead. Also, the (unused) buffer_mb` parameter is gone.
- Sparse estimators in the
sgdmodule use dense parameter vector
`coef_ instead of sparse_coef_`. This significantly improves test time performance.
- The
covariancemodule now has a robust estimator of
covariance, the Minimum Covariance Determinant estimator.
- Cluster evaluation metrics in
~sklearn.metrics.clusterhave been refactored
but the changes are backwards compatible. They have been moved to the metrics.cluster.supervised, along with metrics.cluster.unsupervised which contains the Silhouette Coefficient.
- The `
permutation_test_score` function now behaves the same way as
`cross_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 makes it more intuitive to use with sparse matrix data.
- The functions used for sparse coding, `
sparse_encode` and
`sparse_encode_parallel have been combined into ~sklearn.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_file should 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 renamed
~sklearn.utils.extmath.randomized_svd and 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 <VirgileFritsch> - 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
September 21, 2011
scikit-learn 0.9 was released on September 2011, three months after the 0.8 release and includes the new modules manifold, 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 <https://developers.google.com/open-source/gsoc>_ program.
:target: ../auto_examples/manifold/plot_compare_methods.html
:target: ../auto_examples/linear_model/plot_omp.html
:target: ../auto_examples/decomposition/plot_kernel_pca.html
<div style="text-align: center; margin: 0px 0 -5px 0;">
</div>
| center-div | banner2 | banner1 | banner3 | end-div |
Changelog
- New
manifoldmodule byJake Vanderplas_ and
Fabian Pedregosa_.
- New
Dirichlet Process <dirichlet_process>Gaussian Mixture
Model by Alexandre Passos_
neighborsmodule refactoring byJake 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.
SparsePCAbyVlad Niculae_,Gael Varoquaux_ and
Alexandre Gramfort_
- Printing an estimator now behaves independently of architectures
and Python version thanks to Jean Kossaifi <JeanKossaifi>.
Loader for libsvm/svmlight format <libsvm_loader>by
Mathieu Blondel_ and Lars Buitinck_
- Documentation improvements: thumbnails in
example gallery by Fabian Pedregosa_.
- Important bugfixes in
svmmodule (segfaults, bad
performance) by Fabian Pedregosa_.
- Added
multinomial_naive_bayesandbernoulli_naive_bayes
by Lars Buitinck_
- Text feature extraction optimizations by Lars Buitinck
- Chi-Square feature selection
(feature_selection.chi2) by Lars Buitinck_.
sample_generatorsmodule refactoring byGilles Louppe_
multiclassbyMathieu 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,ShuffleSplit` and 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 Pursuit <linear_model.OrthogonalMatchingPursuit>byVlad Niculae_
- Added 2D-patch extractor utilities in the
feature_extractionmodule byVlad Niculae_
- Implementation of
~linear_model.LassoLarsCV
(cross-validated Lasso solver using the Lars algorithm) and ~linear_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_distances
and metrics.pairwise.pairwise_kernels by Robert Layton
Mini-Batch K-Means <cluster.MiniBatchKMeans>by Nelle Varoquaux and Peter Prettenhofer.
- mldata utilities by Pietro Berkes.
olivetti_faces_datasetbyDavid 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 a `scikits.learn` package 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 `
fit` arguments:
instead all parameters must be only be passed as constructor arguments or using the now public `set_params method inherited from ~base.BaseEstimator`.
Some estimators can still accept keyword arguments on the `fit but this is restricted to data-dependent values (e.g. a Gram matrix or an affinity matrix that are precomputed from the X` data matrix.
- The `
cross_valpackage has been renamed tocross_validation`
although there is also a `cross_val` package 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_func` argument of the
`sklearn.cross_validation.cross_val_score function is now expected to accept y_test and y_predicted as only arguments for classification and regression tasks or X_test` for unsupervised estimators.
- `
gamma` parameter for support vector machine algorithms is set
to `1 / n_features by default, instead of 1 / n_samples`.
- The `
sklearn.hmm` has 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.neighbors` has been made into a submodule. The two previously
available estimators, `NeighborsClassifier and NeighborsRegressor have been marked as deprecated. Their functionality has been divided among five new classes: NearestNeighbors for unsupervised neighbors searches, KNeighborsClassifier & RadiusNeighborsClassifier for supervised classification problems, and KNeighborsRegressor & RadiusNeighborsRegressor` for supervised regression problems.
- `
sklearn.ball_tree.BallTree` has been moved to
`sklearn.neighbors.BallTree`. Using the former will generate a warning.
- `
sklearn.linear_model.LARS()` and related classes (LassoLARS,
LassoLARSCV, etc.) have been renamed to `sklearn.linear_model.Lars()`.
- All distance metrics and kernels in `
sklearn.metrics.pairwise` now 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 parameter `sum_over_features to False`.
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 <JeanKossaifi> - 30 Conrad Lee
- 22 Pietro Berkes
- 18 andy
- 17 David Warde-Farley
- 12 Brian Holt
- 11 Robert
- 8 Amit Aides
- 8
Virgile Fritsch <VirgileFritsch> - 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
May 11, 2011
scikit-learn 0.8 was released on May 2011, one month after the first "international" scikit-learn coding sprint <https://github.com/scikit-learn/scikit-learn/wiki/Upcoming-events>_ and is marked by the inclusion of important modules: hierarchical_clustering, cross_decomposition, NMF, initial support for Python 3 and by important enhancements and bug fixes.
Changelog
Several new modules were introduced during this release:
- New
hierarchical_clusteringmodule by Vincent Michel,
Bertrand Thirion_, Alexandre Gramfort_ and Gael Varoquaux_.
kernel_pcaimplementation byMathieu Blondel_
labeled_faces_in_the_wild_datasetbyOlivier Grisel_.
- New
cross_decompositionmodule byEdouard Duchesnay_.
NMFmoduleVlad Niculae_
- Implementation of the
oracle_approximating_shrinkagealgorithm by
Virgile Fritsch <VirgileFritsch> in the covariance module.
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 byOlivier Grisel_.
- Guide
performance-howtobyOlivier 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_proba` in
~discriminant_analysis.LinearDiscriminantAnalysis By Mathieu Blondel_.
- Refactoring in the
svmmodule and bug fixes byFabian 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 <VirgileFritsch> - 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
March 2, 2011
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 were 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.NeighborsClassifierand
neighbors.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 <https://github.com/scikit-learn/scikit-learn/wiki/Neighbors-working-notes>_ for more information [Fabian Pedregosa_].
- Documentation improvements: Added
pca.RandomizedPCAand
~linear_model.LogisticRegression to 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.LinearSVC or ~linear_model.LogisticRegression [Fabian Pedregosa_].
- Performance and API improvements to
metrics.pairwise.euclidean_distances and to pca.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
December 21, 2010
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
<https://scikit-learn.org/stable/modules/sgd.html>`_ 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 sphx_glr_auto_examples_svm_plot_weighted_samples.py for an example).
- New
gaussian_processmodule 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
(~model_selection.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: sphx_glr_auto_examples_applications_plot_face_recognition.py, sphx_glr_auto_examples_applications_plot_species_distribution_modeling.py, sphx_glr_auto_examples_applications_plot_wikipedia_principal_eigenvector.py and 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.LogisticRegression model.
- 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
October 11, 2010
Changelog
New classes
- Support for sparse matrices in some classifiers of modules
`svm and linear_model (see svm.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.Lars and ~linear_model.LassoLars.
- New Hidden Markov Models module (see classes
hmm.GaussianHMM, hmm.MultinomialHMM, hmm.GMMHMM)
- New module feature_extraction (see :ref:`class reference
<feature_extraction_ref>`)
- 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 <https://scikit-learn.org/stable/modules/svm.html>_ and the complete class reference <https://scikit-learn.org/stable/modules/classes.html>_.
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:
`sphx_glr_auto_examples_mlcomp_sparse_document_classification.py (since removed) and sphx_glr_auto_examples_text_plot_document_classification_20newsgroups.py`
- Many more examples. `See here
<https://scikit-learn.org/stable/auto_examples/index.html>`_ 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
August 26, 2010
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.