[scikit-learn] annotated tag debian/0.17.0_b1-1 created (now 8b05984)
Andreas Tille
tille at debian.org
Wed Dec 28 13:11:07 UTC 2016
This is an automated email from the git hooks/post-receive script.
tille pushed a change to annotated tag debian/0.17.0_b1-1
in repository scikit-learn.
at 8b05984 (tag)
tagging 918005fd5441650ae4a49b510bcabff69ae898bf (commit)
replaces debian/0.16.1-2
tagged by Yaroslav Halchenko
on Tue Oct 27 09:40:38 2015 -0400
- Log -----------------------------------------------------------------
scikit-learn Debian release 0.17.0~b1-1
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iEYEABECAAYFAlYvftYACgkQjRFFY3XAJMgWOACg0G5FBXF05f4+uX2wos5z6BSc
e4EAnjDFeNT1sApfzsSJM1bL0aNgJxM5
=vTY/
-----END PGP SIGNATURE-----
= (3):
add polynomial features to user guide
add connection to kernel method
fix SVM can be tricked into running proba() #4791
Aaron Schumacher (1):
typo: 'precomputed' should be in quotes
Adithya Ganesh (5):
Added check to ensure that NumPy and SciPy meet minimum version requirements (NumPy >= 1.6.1, SciPy >= 0.9)
Amended error message to remain valid when NumPy/SciPy are installed but out of date
Added reference to SciPy/NumPy min_version vars in ImportError
Made setup.py pep8 compliant
Added output of out-of-date SciPy/NumPy versions to setup.py, if detected
Alexandre Gramfort (60):
Merge pull request #4359 from josephlewis42/master
Merge pull request #4364 from amueller/appveyor_badge
Merge pull request #4395 from akitty/patch-1
FIX : allow NaN in input of calibration if estimator handles it
Merge pull request #4476 from CaMeLCa5e/master
Merge pull request #4514 from sinhrks/missing_all
Merge pull request #4519 from ClimbsRocks/patch-1
Merge pull request #4516 from amueller/distance_function_docstring
Merge pull request #4581 from kno10/patch-1
Merge pull request #4596 from ibayer/doc_train_test_split
Merge pull request #4600 from jfraj/bug_20newsgroup
Merge pull request #4647 from lesteve/better-appveyor-badge
Merge pull request #4662 from GaelVaroquaux/infonea_testimonial
Merge pull request #4528 from amueller/rfe_feature_importance_test
Merge pull request #4676 from zacstewart/master
Merge pull request #4161 from rasbt/ensemble-classifier
Merge pull request #4734 from ajschumacher/patch-6
Merge pull request #4679 from TomDLT/newton_cg
Merge pull request #4753 from TomDLT/newton_cg
Merge pull request #4762 from trevorstephens/gridsearch_unused_params
Merge pull request #4778 from untom/fixname
Merge pull request #4777 from TomDLT/liblinear_randomstate
Merge pull request #4792 from amueller/coveralls-badge
Merge pull request #4810 from mblondel/cite_api_paper
Merge pull request #4818 from amueller/sgd_decision_function_deprecation
Merge pull request #4829 from rvraghav93/remove_deprecated_check_cv
Merge pull request #4833 from tw991/feature
Merge pull request #4763 from sonnyhu/RidgeCV_slice_sampleweight
Merge pull request #4835 from tw991/feature2
Merge pull request #4847 from TomDLT/doc_dev
Merge pull request #4857 from christophebourguignat/master
Merge pull request #4861 from ltiao/patch-1
Merge pull request #4562 from hsuantien/emptylabelset
Merge pull request #4894 from tw991/rf
Merge pull request #4893 from jmschrei/cca_stability
Merge pull request #4918 from ejcaropr/doc-fix
Merge pull request #4977 from edsonduarte1990/master
Merge pull request #5001 from hlin117/svm-documentation
Merge pull request #4933 from banilo/regr_score_docs
Merge pull request #5011 from TomDLT/rcv1_order
Merge pull request #5025 from jakirkham/fix_typo
cosmit in plot_cv_predict
Merge pull request #4960 from mkrump/lb_encoder
Merge pull request #5092 from kazemakase/fix-lda-covariance
typos
Merge pull request #5126 from rasbt/rbf
Merge pull request #5143 from Eric89GXL/fix-r2
Merge pull request #5157 from amueller/ridge_intercept_docs
Merge pull request #5131 from michigraber/nonnegative-lars
pep8 + simplify positive test
unused import
Merge pull request #5216 from JPFrancoia/master
Merge pull request #5258 from amueller/docfixes
Merge pull request #5283 from TomDLT/remove_warnings
Merge pull request #5335 from rrohan/current
misc + pep8
Merge pull request #5336 from giorgiop/pca-warning
Merge pull request #5363 from Naereen/patch-1
Merge pull request #5365 from Naereen/patch-2
DOC : update docstring of l1_ratio for E-Net CV classes
Alexey Grigorev (1):
printing top terms even if LSI is used
Ali Baharev (1):
Typo: PCA is not the abbreviation of Probablisitic
Allen Riddell (1):
DOC: Fix typo in CalibratedClassifierCV
Ando Saabas (1):
Store tree node info for all nodes, instead of just leaves
Andreas Mueller (207):
Move dev branch to 0.17
add appveyor badge to alert us of failures
Merge pull request #4363 from arimbr/patch-1
add whatsnew entry for GaussianNB sample weights
make landscape.io much more useful
Merge pull request #4331 from martin0258/uni-feature-doc-seealso
Merge pull request #4335 from martin0258/dpgmm-lowerbound-bug
Merge pull request #4378 from dan-blanchard/patch-1
remove some style errors in bench code.
Merge pull request #4381 from amueller/style_errors_b
Merge pull request #4383 from saketkc/maint
Pass the appropriate include_self argument to kneighbors_graph everywhere.
DOC make defaults more explicit in text feature extraction.
Merge pull request #4392 from sinhrks/setparams
Merge pull request #4390 from Barmaley-exe/tests-fix
fix ompcv on old scipy versions
remove deprecated stuff from 0.17
remove cross_validation.Bootstrap
Merge pull request #4411 from ragv/maint_remove_n_iterations
Merge pull request #4412 from ragv/maint_remove_n_iterations_2
avoid NaN in CCA
FIX Minor shape fixes in PLS
make cross_decomposition pass common tests
deal with y lists
DOC fix n_jobs docs in KMeans as in 0a611193b12900dbc11f3dae4448809364161bb2.
DOC fix typo in output shape of fetch_lfw_pairs (and minor additions)
FIX LDA(solver="lsqr"): make sure the right error is raised on transform.
DOC fix link description to setuptools in devdocs.
COSMIT missing newlines in metrics test
Merge pull request #4436 from ogrisel/rebased-pr-3747
fix setting description in 20newsgroups
cleaned up input validation, added tests for copy keyword.
ENH minor fixes / simplifications for PLS input validation, add test for error.
DOC minor fixes in formatting, don't use deprecated n_components in Agglomerative
DOC/MAINT remove deleted cluster.Ward from references.
Merge pull request #4451 from ragv/travis_ignore_docstring
Fixes to sparse decision_function, rebase fixes.
DOC added highlights to whatsnew
DOC / website add 0.16 release to "news".
Merge pull request #4462 from xuewei4d/deprecate_estimator_params_docstring
Merge pull request #4471 from baharev/small-typo
FIX don't raise memory error in ledoit wolf
Merge pull request #4428 from amueller/ledoit_wolf_shrinkage_fix
Add tags to classifiers and regressors to identify them as such.
COSMIT use consistent shape description in docstring.
Merge pull request #4464 from vortex-ape/logregcv
Merge pull request #4456 from cgohlke/patch-1
Merge pull request #4356 from vortex-ape/dict_vectorizer
Merge pull request #4484 from vortex-ape/remove_array2d
Merge pull request #4496 from vmichel/rfe_feature_importances
DOC adding clusterer tag to dev docs.
Merge pull request #4418 from amueller/classifier_regressor_tags
DOC try to clarify pairwise_distances docstring for Y.
DOC fix link in text extraction narrative
Merge pull request #4506 from mheilman/master
DOC GradientBoostingX.estimators_ docstring shape.
FIX make CalibratedClassifierCV deterministic by default.
Merge pull request #4537 from amueller/callibrated_classifier_cv_linearsvc_seed
FIX be robust to columns name dtype and also to dataframes that hold dtype=object.
Merge pull request #4531 from zhaipro/fix
Merge pull request #4561 from mspacek/patch-1
don't to input validation in each tree for RandomForest.predict
Misc trying to make k_neighbors warning even more explicit (before it didn't say that %d was n_samples)/
add 0.16.1 bugfixes to whatsnew.
Merge pull request #4467 from giantoak/param_invariance
Merge pull request #4534 from xuewei4d/optimize_rfecv
DOC fix libsvm docstring consistency
Merge pull request #4636 from scls19fr/scls19fr-Update_confusion_matrix_docstring
Merge pull request #4631 from Aerlinger/documentation_improvements
ENH add high-level estimator-validation function
check output, not error on sparse data handling
fix warning message by not using assert_raise_message, which adds escaping.
add comments to tests
FIX make backport of assert_raises_regex raise the same error as the python3 version.
FIX make rfe feature importance test deterministic.
Merge pull request #4640 from dougalsutherland/patch-1
Merge pull request #4526 from vortex-ape/if_matplotlib
FIX pass percentiles to partial_dependence in plotting.
Merge pull request #4593 from clorenz7/gmm_fit_predict
Merge pull request #4659 from MAnyKey/sampler-samples
fix minor documentation link issues. looks more complicated than it is as I replaced tabs with spaces
FIX work-around for read only dataframes
Merge pull request #4678 from amueller/pandas_safe_indexing_fix
ENH make pipeline.named_steps a property, fix pipeline.named_steps doctest
DOC whatsnew for unaveraged multi-output regression metrics.
Merge pull request #4438 from mbatchkarov/stratified_train_test_split
DOC add whats_new for stratified train_test_split
DOC add random_state to parameter docstring in gradient boosting
ENH don't use mutable defaults in RidgeCV.
FIX add missing numpy imports to VotingClassifier examples, pep8 fixes.
COSMIT remove unnecessary float casting from tomography example
TST don't check that warning was raised as whether a copy needs to be made or not depends on the pandas version. Version 0.12 and 0.13 apparently don't use cython here, and >0.17 will have a fix.
FIX / TST raise error when init shape doesn't match n_clusters in KMeans, check for sensible errors.
DOC fix sample_without_replacement docstring. Closes #4699.
Merge pull request #4694 from betatim/check-fit-returns-self
Merge pull request #4702 from betatim/online-learning-batch-size
Doc sphinx fix Return->Returns in label_ranking_loss
Merge pull request #4727 from Spikhalskiy/truncated-svd-doc-T
FIX make random_choice_csc and therefor DummyClassifier deterministic.
ENH support for sparse coef_ in ovr classifier
TST test that all default arguments are not mutable
DOC some fixes for the EnsembleClassifier
Merge pull request #4645 from TomDLT/astype_fix
FIX / TST make cross_val_predict work on lists, test pass-through and compatibility with various input types.
FIX ransac output shape, add test for regressor output shapes
don't warn about multi-output deprecation if not using multi-output
Merge pull request #4743 from ogrisel/fix-check-finite-gram-omp
Merge pull request #4684 from ogrisel/readonly-lars-path
Merge pull request #4749 from ogrisel/travis-no-sudo
FIX n_jobs slicing bug in dict learning, add input validation to gen_even_slices
Merge pull request #4716 from TomDLT/notfitted
Merge pull request #4125 from untom/RobustScaler
Merge pull request #4761 from tiagofrepereira2012/fix_lle
DOC added some bug fixes to whats_new.rst
TST/COSMIT remove nose call boilerplate
Fix RFE / RFECV estimator tags
DOC Add coveralls badge to Readme
Merge pull request #4680 from amueller/pipeline_named_steps_fix
Merge pull request #4431 from vortex-ape/deprecate_func
Use more natural class_weight="auto" heuristic
FIX add class_weight="balanced_subsample" to the forests to keep backward compatibility to 0.16
Cosmit readability improvements and better whatsnew.
Merge pull request #4785 from amueller/no_noise_boilerplate
float division
Merge pull request #4786 from amueller/estimator_tags_bug
DOC adding backlinks to docstrings
Merge pull request #4723 from amueller/backlinks
DOC minor sphinx fixes
Fix deprecation of decision function in SGD
ENH refactor OVO decision function, use it in SVC for sklearn-like decision_function shape
TST set missing random state in test
ENH minor fixes to the tests, don't raise as many warnings in the test suite
Merge pull request #4819 from sornars/svcgamma
Merge pull request #4832 from kronosapiens/update-docs
Better docstring for PCA, closes #3322.
Merge pull request #4838 from trevorstephens/ridge_sw
Merge pull request #4796 from joshloyal/standard-scaler-std-doc
DOC remove linewidth in pca/ica example
Merge pull request #4854 from opahk/master
Merge pull request #4849 from jnothman/contrib-link
Merge pull request #4868 from rasbt/silhouette
Merge pull request #4870 from rasbt/decisiontreeregressor
Merge pull request #4836 from amueller/pca_docstring
Merge pull request #4947 from zyrikby/patch-1
Merge pull request #4953 from mrphilroth/issue4940
Merge pull request #4783 from tokoroten/decrease_randomforest_memory
Merge pull request #4962 from vinc456/unexpected-facial-recognition
Merge pull request #4915 from untom/minmax_scale
Merge pull request #4971 from dotsdl/issue-4724
add whatsnew entry for fix of penalty passing in LogisticRegressionCV
remove links to old docs
Make more clear that adding algorithms is not the preferred way to start contributing.
Merge pull request #5038 from jnothman/textdoc
Merge pull request #5042 from amueller/contributing_addition
fix shape of thetaL, thetaU, theta0 in gp
Merge pull request #5048 from marktab/fixlfw
Merge pull request #4965 from mrphilroth/4922
DOC fixed short underline
FIX make sure y is 1d in svm regressor, add common test for regressor shapes.
fix TheilSenRegressor input validation, LassoLarsIC input validation
added handling of 2ndim y in regressors to whatsnew.
Merge pull request #5057 from amueller/fix_2d_y_svm_regressor
Merge pull request #4775 from arthurmensch/cd_fast_readonly_array_brainfix
Merge pull request #5052 from larsmans/kappa
force y to be numeric in kernel ridge.
TST Fix: made the test for y 1d conversion warning more robust.
Merge pull request #5077 from rvraghav93/fix_import_coor_des
ENH: Renames CallableTransformer -> FunctionTransformer.
DOC added whatsnew for k-means fix
FIX be robust to obscure callables that are false.
Add common test that transformers don't change n_samples.
Fix memory access error in OneClassSVM
Merge pull request #5093 from amueller/one_class_svm_memory_fix
Merge pull request #5117 from kylerbrown/patch-1
Merge pull request #5120 from larsmans/faster-lda
DOC mention intercept_ attribute in ridge docstring.
test for accepted sparse matrices
Merge pull request #5037 from betatim/tree-feature-transform
Merge pull request #5195 from jnothman/example_latex
Merge pull request #5218 from mooz/patch-1
Merge pull request #5199 from ogrisel/joblib-0.9.0b4
Merge pull request #5200 from jackzhang84/add_dump_svmlight_file_comment
Updated whatsnew for sprint and LatentDirichletAllocation, add some missing author links
fix rebase mistake
add necessary blas files.
Minor fixes, remove transform for now.
minor fixes to the doc build
Merge pull request #5263 from hlin117/check_X_y-docs
Merge pull request #5238 from rvraghav93/cvdoc
Merge pull request #4887 from amueller/tsne_fixes
Merge pull request #4852 from TomDLT/nmf
add future warning to pipeline.inverse_transform with 1d X.
Merge pull request #5294 from trevorstephens/lda_qda_fit_params
Merge pull request #4955 from copyconstructor/preserve-vocab-ordering
Merge pull request #5349 from jakevdp/naive-bayes-scale
DOC trying to make GridSearchCV docs more accurate... not sure if better
Fixes #4455.
Merge pull request #5104 from giorgiop/scaler-partial-fit
Merge pull request #5400 from kwgoodman/doctypo
DOC formatting fixes for laplacian kernel
Fixed warnings for DataDimensionalityWarning, decision_function and decision_function_shape.
Merge pull request #5410 from GaelVaroquaux/fix_doc_gcv
Merge pull request #5398 from rvraghav93/fix_for_numpy_10
Merge pull request #5409 from lesteve/fix-plot-lle-digits
MAINT don't print things in testing.
FIX Don't compare arrays to strings!!!!
DOC cosmit I find it confusing to say that fit resets the estimator, at it always does that.
Andrew Lamb (5):
Fix check in `compute_class_weight`.
Fix assertion for Python 3, use `assert_raise_message`.
Change formatting of `assert_raise_message`.
Fix SGD partial_fit multiclass w/ average.
Added entry to `doc/whats_new.rst`.
Anish Shah (1):
[issue #5043] fix documentation of callable kernel
Ankur Ankan (1):
fixes bug in oob_score when X is sparse.csc matrix [refs #4744]
Anthony Erlinger (5):
Minor docstring fixes to svm/classes.py
Improve documentation on Diabetes dataset
Typo fix in diabetes documentation :Aex: -> :Sex:
Improve wording of SVM `crammer_singer` doc following PR feedback.
Clarify documentation on feature scaling for diabetes data
Ari Rouvinen (1):
DOC Fix NMF inconsistency and broken links
Arnaud Joly (21):
FIX numpy deprecation warning from unsafe type comparison
ENH add a prior strategy for the dummy estimator
Update what's new
FIX raise error properly when n_features differ in fit and apply
Factorize input validation
ENH Factor out validation of X for apply/predict in forest
FIX properly raised not fitted error
FIX use the new notfittederror in feature_importances_
DOC use same error message in decision tree and random forest
ENH add ranking_loss multilabel ranking metric
FIX raise ValueError if sample weight are passed but unsupported by the base estimator
Update what's new
DOC document missing attributes
FIX raises memory error in depth first builder
Merge pull request #4932 from rasbt/randomforest
Optimize MSE criterion
ENH Faster tree-based methods by implementing reverse update of criterion
Merge pull request #5230 from jmschrei/_tree_split
FIX unstable test due to bootstrap and unset random state
Merge pull request #5252 from jmschrei/presort
DOC FIX missing minus in Shannon's entropy formula
Arthur Mensch (3):
Bugfix : type in cd changed for read only memmap compatibility
Added check_input in Lasso for DictLearning performance
Fix sparse_encode input checks
Barmaley.exe (4):
DOC Adds SVR description to the tutorial
TST Adds sanity check for SVR's decision_function
DOC A small change to the SVM's tutorial
TST Fixes #4386
Boyuan Deng (3):
Use immutable default arguments throughtout repo
Link to functions and fix typos in dataset docs
Fix class reference in twenty_newsgroups.rst
Brett Naul (2):
Added/improved GraphLasso tests, per #4134
Add enet_tol parameter to GraphLasso class/methods
Brian McFee (1):
ENH: added LabelShuffleSplit cv iterator
Buddha Prakash (7):
ENH: split LDA's n_iter_ into n_iter_ and n_batch_iter_
Add check for sparse prediction in cross_val_predict
Use a single vstack for concatenating all blocks in prediction matrix
Use Inverted locations to reorder the predictions
Remove redundant p variable
Add test to check sparse predictions in cross_val_predict
Fix minor indentation issues
CJ Carey (4):
WIP: adding 'max' normalizer to normalize()
TST: covering norm='max' branches of normalize()
DOC: updating Normalizer docstring for norm='max'
Fixing sparse max for older scipy
Chi Zhang (2):
added comments for y in dump_svmlight_file() method
changed shape parameter
Chih-Wei Chang (1):
Add multilabel support for dump_svmlight_file.
Christof Angermueller (1):
Add conventions section to userguide
Christoph Gohlke (1):
Fix signature not compatible with previous declaration
Christophe Bourguignat (5):
DOC document missing attributes
DOC Updated documentation for cv parameter
DOC Updated documentation for cv parameter (issue #4533)
DOC Updated documentation for cv parameter (issue #4533)
[DOC] Precision on random_state in KFold() doc
Christopher Erick Moody (1):
FEAT Barnes-Hut t-SNE
Chyi-Kwei Yau (31):
add OnlineLDA model and test
don't use sparse.block_diag since it is new in scipy 0.11.0
fix typos and imporve documentation
fix typo & improve doc 2
testing code improvement
move input validation and variable assignment to fit method
fix _approx_bound with subsampling
native dense input support
rename variable and minor pep8 fix
clean up test and add error check in preplexity method
remove duplicate code in _approx_bound method
move check_non_negative to utils/validation.py
fix params and import test coverage
update example and change total_samples in fit method
add LDA user guide
add LDA to classes.rst
random initialize in e-step
change LDA example to script
change doc, fix n_jobs, and remove fit_transform
merge NMF and LDA example
improve LDA doc
fix typo and improve doc
use n_features_
remove self.n_features_
remove main in test
make dimension mismatch err msg more explicit
change all rng to random_state
fix feature_names
improve document in _online_lda.pyx
use logsumexp
remove dirichlet_component_ variable
Cindy Sridharan (3):
vocabulary of type set now coerced to list to preserve iteration ordering after serialization
use sklearn.utils.random
removed list around sorted
Clyde-fare (1):
Added laplacian kernel
Cory Lorenz (2):
Fix Float Resolution Bug on Gaussian Process Test
Add a fit_predict method for the GMM classes
Dan Blanchard (1):
Vainly add link to my GH profile in What's New.
Daniel Galvez (6):
Make apply method of trees public. Added test for concistency with private method.
Added docstring
Added example demonstrating tree.apply
Added indentation to docstring
Removed cruft
Added tests of apply() for valid and invalid inputs. Fixed style.
Daniel Kronovet (3):
fixed typos and update formatting in doc/developers/index
additional typos, formatting, and links for doc/developers/index and /utilities
Fixed factual error in descriptino of PR flow
Danny Sullivan (1):
Adding Implementation of SAG
Data1010 (2):
fix #5139
DOC FAQ on loading data as numpy arrays
David (1):
[Pipeline] add named_steps attribute documentation #4482
David D Lowe (3):
Add example for precision_recall_fscore_support
Shrink the doc for precision_recall_fscore_support
Shrink precision_recall_fscore_support's doc again
David Dotson (2):
Closes #4614
Added list of multi-label supporting classifiers.
Dmitry Spikhalskiy (1):
Fix mistake in TruncatedSVD expalation in docs
Donne Martin (1):
Fix #4978: Typo in CONTRIBUTING.md
Dougal J. Sutherland (1):
ForestClassifier.predict docstring correction
Dougal Sutherland (1):
support X_norm_squared in euclidean_distances
Eduardo Caro (1):
Change name of variable to be consistent with dataset
Eric Larson (1):
FIX: Fix for changing mode
Eric Martin (1):
Directly compute polynomial features.
Erich Schubert (1):
Update whats_new.rst
Fernando Carrillo (1):
Updated link. Original link to ZhangIJCV06 paper "Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive Study" was dead. Replaced all occurances of "http://eprints.pascal-network.org/archive/00002309/01/Zhang06-IJCV.pdf" with "http://research.microsoft.com/en-us/um/people/manik/projects/trade-off/papers/ZhangIJCV06.pdf"
Frank C. Eckert (1):
Fix typo in model evaluation documentation
Frank Zalkow (9):
added a string for FriedmanMSE (instead impurity) when exporting a dot file
missed elif
now uses isinstance and keeps original name (FriedmanMSE)
added a test and reverted string to friedman_mse
corrected test: criterion name has to be only in those nodes where there is "samples"
took if clause into regex
implemented improvement suggestions
added newline (for pep8) and reverted to regex solution due to 0.0/-0.0 problem on windows
fixes
Gael Varoquaux (58):
DOC: stock_market example: follow change of symbol
Merge pull request #4401 from ogrisel/rebased-pr-4157
Merge pull request #4433 from gwulfs/master
Merge pull request #4389 from vortex-ape/move_test
MAINT: update mailmap
Merge pull request #4505 from jmetzen/isotonic_make_unique_fix
Merge pull request #4563 from mspacek/patch-2
Merge pull request #4560 from ogrisel/shuffle-passthrough
Merge pull request #4588 from amueller/random_forest_predict_check
Merge pull request #4541 from amueller/robust_input_dtype_check
Merge pull request #4595 from sseg/bicluster-example
Merge pull request #4592 from clorenz7/bugfix/gauss_proc_eps_issue
Merge pull request #4434 from xuewei4d/catch_LinAlgError_gmm
DOC: add a testimonial by infonea
Merge pull request #4653 from ssaeger/issue_4633
Merge pull request #4681 from trevorstephens/gplearn-link
Merge pull request #4665 from kianho/plot-ensemble-oob-trace
Merge pull request #4695 from perimosocordiae/patch-1
Merge pull request #4686 from bnaul/graph_lasso_tests
Merge pull request #4663 from Titan-C/sidebar
Merge pull request #4347 from amueller/class_weight_auto
Merge pull request #4952 from stephen-hoover/fix_nmf_init_doc
Merge pull request #4972 from ogrisel/doc-whatsnew-joblib-0.9.0b2
Merge pull request #5018 from ltiao/patch-1
Merge pull request #5016 from lesteve/update-joblib-to-0.9.0b3
Merge pull request #5049 from amueller/gp_reshape_fixes
Merge pull request #5047 from amueller/scipy_0.16_fixes
Merge pull request #5072 from amueller/fix_fun_conversion_warning
DOC: minor cosmetric to ROC example
Merge branch 'master' into pr_3651
Merge branch 'master' into pr_4009
Merge branch 'master' into pr_4009
Merge pull request #5187 from GaelVaroquaux/pr_4009
Merge pull request #4253 from Flimm/doc-precision_recall_fscore_support
DOC: update whats_new
TST: Use a local random state
Merge pull request #4104 from maheshakya/johnson-lindenstrauss_bound_example_fix
Merge pull request #4312 from xuewei4d/friendly_error_on_kneighbors
TST: No n_jobs=-1 in the test
Merge pull request #5188 from GaelVaroquaux/pr_4090
Merge pull request #5184 from christophebourguignat/master
Merge pull request #4767 from trevorstephens/passive-aggressive_cw
Merge pull request #4779 from martinosorb/parallel-ms
Merge pull request #4881 from sonnyhu/weighted_least_squares
Merge pull request #5189 from GaelVaroquaux/pr_4779
Merge pull request #5182 from MechCoder/predict_proba_fix
Merge pull request #2459 from dougalsutherland/euclidean_distances
Merge pull request #5194 from christophebourguignat/master
Merge pull request #5201 from christophebourguignat/master
Merge pull request #5289 from carrillo/master
Merge pull request #5253 from ogrisel/online-lda-with-parallel-context-manager
Merge pull request #5301 from ogrisel/python-3.5-inspect-deprecation
Merge pull request #5311 from jereze/master
Merge pull request #5317 from hlin117/verbose-sparse-encode
Merge pull request #5334 from ogrisel/better-ensure-min-error-message
Merge pull request #5332 from larsmans/nmf-cleanup
Merge pull request #5346 from Naereen/patch-1
DOC: correct docstring
Ganiev Ibraim (4):
OneClassSVM sparsity regression test added
deleted useless assert
Removed unnecessary class checks
Style fix
Gilles Louppe (42):
Merge pull request #4489 from betatim/rf-oob-docs
ENH: public apply method for decision trees
ENH: raise NotFittedError instead of ValueError
Update what's new
DOC: better documentation tree.apply
Merge pull request #4488 from glouppe/tree-apply
Merge pull request #4570 from thedrow/topic/call-enumerate-on-empty-clusters-only
Merge pull request #4530 from amueller/gbrt_estimators_docstring
Merge pull request #4642 from arjoly/error-bagging-sw
COSMIT: VotingClassifier
Merge pull request #4692 from kianho/plot-ensemble-oob-trace
Merge pull request #3735 from trevorstephens/pretty-decision-trees
Merge pull request #4698 from trevorstephens/doc-export_graphviz
Merge pull request #4958 from rasbt/treeregressor
Merge pull request #4979 from donnemartin/master
Merge pull request #4954 from ankurankan/fix/4744
FIX: ensure that get_params returns VotingClassifier own params
Merge pull request #5010 from jmschrei/_tree_doc
DOC: fixes to _tree.pyx
DOC: add cross_validation.LabelShuffleSplit to classes.rst
Merge pull request #5162 from ogrisel/faq-freeze
ENH: rename to LabelKFold
COSMIT: variable names, documentation, etc
ENH: remove unnecessary assignments
Merge pull request #5220 from arjoly/fast-reverse-update
Merge pull request #5190 from glouppe/labelkfold
Merge pull request #5223 from mattilyra/allow_sparse_in_baggingclassifier
Merge pull request #5207 from MechCoder/ensemble_sample_weight_bug
Merge pull request #5242 from arjoly/stable-test-2
TEST: stronger tests for variable importances
TEST: use sklearn.fixes.bincount
TEST: take comments into account
TEST: reduce test time, variable name, etc
TEST: check parallel computation
Merge pull request #5261 from glouppe/check-importances
Merge pull request #5278 from jmschrei/_criterion_cleanup
DOC: better docstring for sum_total
Merge pull request #5305 from jmschrei/criterion_patch
FIX: Ensure correct LabelKFold folds when shuffle=True
Merge pull request #5310 from JeanKossaifi/add_repo
Merge pull request #5300 from glouppe/fix-5292
Merge pull request #5347 from ariddell/patch-3
Giorgio Patrini (3):
remove numpy's RuntimeWarning from corner case of PCA.fit
partial_fit for scalers
BUG: reset internal state of scaler before fitting
Graham Clenaghan (1):
Implement dropping of suboptimal thresholds in roc_curve
Gryllos Prokopis (3):
Add more extensions and related packages to related_projects.rst
Add some newline characters to respect the 80 characters rule
Fixed minor typo
Henry Lin (5):
fixed newline errors in documentation for SVC
fixed newline errors in documentation for NuSVC
fixed newline errors in documentation for LinearSVC and LinearSVR
check_X_y: Added mention that function returns y array
#5309: Added verbosity argument to sparse_encode, verbosity for LassoLars and Lasso
Hsuan-Tien Lin (8):
a simple fix to reading empty label sets in the multilabel case
cython compiled from pyx
directly set target to empty list instead of empty string
given that line_parts has been copied to features, parsing should be based on features rather than line_parts
cython the pyx
use COLON to avoid byte buffer error
add the empty label test case
add another test case with empty labelset
Immanuel Bayer (2):
DOC use standart X, y notation
remove order checks for sparse matrices
Ishank Gulati (1):
Breast cancer dataset added
Jack Martin (2):
Added import statement for Lasso example
added import np statements
Jacob Schreiber (16):
Epsilon added to x and y weights in nipals inner loop for numeric stability on windows
Changed addition of epsilon for cases where y_weights is near 0
Generated C file
Merge branch 'master' of https://github.com/scikit-learn/scikit-learn
Documentation added to criterion, dense splitters.
Verbosity reduced, suggestions merged
Responses changed to targets
Responses changed to targets
PEP257 adhered to closer
Final round of revisions
ENH split _tree.pyx into several files
ENH apply method added to Gradient Boosting
ENH criterion file cleaned up
FIX criterion variable names
ENH gbt sparse support
FIX mailmap
Jaidev Deshpande (1):
ENH Add option in cosine_similarity for sparse output
Jake VanderPlas (7):
Merge pull request #5231 from vighneshbirodkar/kmeans_fix
MAINT: make GaussianNB scale-invariant
TST: GaussianNB scale invariance
GaussianNB: use var() rather than std() in epsilon determination
MAINT: minor fixes to GaussianNB epsilon
DOC: add whats_new entry for PR #5349
fix merge conflict
Jan Hendrik Metzen (3):
ENH Adding sample-weight support for Gaussian naive Bayes
TST Tests for sample-weights in Gaussian naive Bayes
FIX _make_unique of _isotonic.pyx uses float64 consistent internally
Jean Kossaifi (2):
Added subject independent KFold
Added link to github repositories in what's new.
Jeremy (1):
Dataiku testimonial
Jiali Mei (1):
a common test to check if classifiers fail when fed regression targets
Joe Jevnik (2):
ENH: Adds CallableTransformer
COMPAT: Makes test_function_transformer py2 compatible.
Joel Nothman (38):
STY remove unnecessary variable/operation
Merge pull request #4555 from ogrisel/fix-astype-copy-true
ENH faster LSHForest: use take rather than fancy indexing
ENH labels parameter in P/R/F may extend or reduce label set
DOC make contributor's guide more prominent on front page
PEP8 for recent merge
DOC backticks in attribute docstrings unnecessaru since #3489
DOC minor tweaks to docs homepage
DOC organise documentation hierarchy / table of contents
DOC links should not always point to dev
FIX avoid memory cost when sampling from large parameter grids
FIX #4902: SV*.predict_proba visible before fit
DOC tweaks for feature_extraction.text
COSMIT prefer loops to repetition in LSHForest benchmark
DOC rearrange related projects
DOC use correct attribute in example
DOC don't show example thumbnails in latex docs
ENH precomputed is now a valid metric for 'auto' neighbors
ENH support precomputed neighbors where query != index
TST precomputed matrix validation in correct place
FIX add _pairwise and test to NeighborsBase
DOC fix up parameter descriptions
FIX broken rebase
TST add test for precomputed metric and X=None at predict
FIX test with invalid input; simplify dbscan precomputed
FIX error introduced during rebase
ENH sparse precomputed distance matrix in DBSCAN
DOC resize images for latex
DOC hide web-specific links from latex build
DOC prefix version headings by "Version"
DOC provide PDF documentation for download
ENH avoid slow CSR fancy indexing variant
DOC comment on obscure syntax
Merge pull request #5288 from jnothman/bicluster-example-speed
Merge pull request #5237 from gclenaghan/remove_thresholds
Merge pull request #5369 from joshloyal/fix_hashing_docstring
Merge pull request #5371 from ylow/sgd_bug_fix
Merge pull request #5402 from lesteve/fix-lshforest-doctest
John Kirkham (1):
DOC: Remove extra "not". [ci skip]
John Wittenauer (2):
Added several related projects.
Added a link to the deep learning software list.
Joseph (1):
Related projects linking to wrong URL
Joshua Loyal (3):
Updated _std docstring in StandardScaler to make internal handling of
ENH Document how to get valid parameters in set_params
correct optional arguments for FeatureHasher
Jungkook Park (1):
Use expit function to compute the probability in ExponentialLoss class.
Kashif Rasul (1):
make_multilabel_classification sparse target
Keith Goodman (1):
DOC wrong default value in docstring
Kian Ho (5):
Initial commit of plot_ensemble_oob.py
Minor amendments to plot_ensemble_oob.py.
Amended plot_ensemble_oob.py according to #4665.
Made more amendments according to PR feedback.
Refined the docstring for plot_ensemble_oob.py.
Konstantin Shmelkov (1):
ENH multioutput regression metrics
Kyler Brown (1):
fixing a typo in feature extraction documentation
Lars Buitinck (39):
ENH optimize DBSCAN by rewriting in Cython
FIX TSNE.fit: didn't return self
DOC typo in TruncatedSVD narrative
Merge pull request #4739 from amueller/regressor_output_shape_test
DOC fix comment in tf-idf: log+1, not log(1+x)
Merge branch 'pr/4714' + pep8 fixes
TST fix failing doctest
Merge pull request #4905 from lesteve/update-joblib-0.9.0b2
Merge pull request #4900 from arjoly/fix-raise-memerror
Merge pull request #4966 from lazywei/multilabel-dump-svmlight
Merge pull request #4937 from pianomania/FeatureHasher_add_example
DOC FeatureHasher takes (finite) numbers as values
Merge pull request #4956 from nealchau/ransacresid
FIX dump_svmlight_file multilabel arg: should be last
ENH: add Cohen's kappa score to metrics
FIX assign_rows_csr: should not zero the entire out array
FIX commit re-cythonized sparsefuncs_fast
TST common metric tests for cohen_kappa_score
DOC expand FunctionTransform docstring
Merge pull request #5059 from amueller/function_transformer_rebase
DOC/MAINT final touches to FunctionTransformer
Merge pull request #5076 from PGryllos/update_doc
Merge pull request #3659 from chyikwei/onlineldavb
COSMIT pep8 online LDA
ENH micro-optimize LatentDirichletAllocation + cosmetics
Merge pull request #5074 from jdwittenauer/master
DOC: happy new year!
DOC: typo in function name
Merge pull request #5050 from AnishShah/issue5043
Merge pull request #5111 from chyikwei/remove-unnecessary-variable
TST: LatentDirichletAllocation behavior on empty documents
ENH: faster LatentDirichletAllocation (~15% on 20news)
Merge pull request #5271 from vortex-ape/lda-typo
ENH: optimize LDA, ~15% faster on one core
DOC: TODO notice in Hungarian algorithm, use SciPy version when released
DOC: optimize Dataiku logo (16kB => 10kB)
BUG shuffle components, not samples in NMF CD
MAINT: better timing and general fixes in NMF/LDA example
COSMIT some refactoring in NMF
Lilian Besson (3):
Fixed typo in the final ".. note::" of naive_bayes.rst
Missing link for inertia (about K-Means)
Fix few typos on links and doi
Loic Esteve (16):
DOC remove sphinx warnings when generating the doc
Use anonymous hyperlink targets
Tweak appveyor badge
FIX DBSCAN fit with precomputed matrix in edge cases
Updated joblib to 0.9.0b2
Update joblib to 0.9.0b3
DOC fix README.rst coveralls badge
MAINT make copy_joblib.sh Python 3 compatible
MAINT update joblib to 0.9.2
FIX non deterministic LSHForest doctest failure
DOC add link to the joblib 0.9.2 release notes
DOC Fix plot_tomography_l1_reconstruction example
MAINT Update the 3rd travis build to latest versions of numpy and scipy
FIX AdaBoostRegressort test failure with numpy 1.10
MAINT add safe_{median|mean} for np 1.10.1
DOC fix missing import in plot_lle_digits example
Louis Tiao (2):
Update plot_ols.py
Filled in missing negation
Maheshakya Wijewardena (1):
Implemented SelectFromModel meta-transformer
Manoj Kumar (24):
[MRG] [BUG] Pass penaly to the final logistic regression fit
[BUG] predict_proba should use the softmax function in the multinomial case
override predict_proba in log_reg
Add non regression test
AdaBoostRegressor should not raise errors if the base_estimator
Add numerically stable softmax function to utils.extmath
Add test for softmax function
Merge pull request #5225 from MechCoder/fix_overflow
[BUG] _init_centroids has an optional x_squared_norms parameter which is not exactly optional
Merge pull request #5262 from andylamb/lamb-fix-class-weight-check
Merge pull request #5284 from tdhopper/patch-1
fix test failures
Remove warm start
Catch filters instead of removing the tests
Added example to depict feature selction using SelectFromModel and Lasso
Minor doc changes and removed _set_threshold and _set_importances
Now a fitted estimator can be passed to SelectFromModel
Add narrative docs and fix examples
Merge SelectFromModel and L1-selection examples
Lasso and ElasticNet should handle non-float dtypes for fit_intercept=False
Merge pull request #4707 from amueller/k_means_init_mismatch
1. Added parameter prefit to pass in a fitted estimator.
Refactor tests
Merge pull request #4242 from MechCoder/select_from_model
MarkTab marktab.net (1):
Update lfw.py
Martin Ku (2):
Fix typo of DPGMM doc
Add "See also" for selectors and scoring funs
Martin Spacek (2):
DOC fix fastica source matrix output shape
Improve FastICA convergence warning message
Masafumi Oyamada (1):
Correct expectation notation in DP-GMM doc
Mathieu Blondel (11):
Support sparse matrices in KernelRidge.
Merge pull request #4713 from amueller/ridge_cv_nonmutable_alphas
Merge pull request #4733 from amueller/fix_csc_random_sampling
Typo.
Add API paper.
Reference ECML PKDD workshop paper.
Merge pull request #4851 from trevorstephens/ridge_no_copy
Merge pull request #4891 from MechCoder/lrcv_bug
chmod -x
Remove shebang and utf-8 header.
Add link to ETL.
Matt Krump (2):
Don't allow negative values for inverse transform, and raise error similar to transform for new labels
Added test
Matti Lyra (2):
Added the option of passing in a sparse X matrix into decision function, plus tests for sparse for all prediction functions.
Changed predict_log_proba so that it accepts sparse matrices.
Maxim Kolganov (2):
remove useless |samples| variable out of ParameterSampler.__iter__()
replace itertools.repeat with six.moves.range
Michael Heilman (2):
fixed reverse sorting issue in label_binarize()
added an additional test case in test_label.test_label_binarize_with_class_order()
Michael Patterson (2):
Changed the number of features from 10 to variable
Now plots based on the number of features in X
Miroslav Batchkarov (4):
[WIP] preliminary implementation of stratified train_test_split
fixed issue when train_size < test_size for stratified
using StratifiedShuffleSplit instead of StratifiedKFold
incorporated amueller's feedback
Nelle Varoquaux (1):
Merge pull request #4420 from amueller/cca_low_rank
Nicolas (1):
FIX make StandardScaler & scale more numerically stable
Nikolay Mayorov (1):
Added multithreading support for kneighbors search
Olivier Grisel (148):
FIX #4358: more tolerant test_scorer_sample_weight for windows
FIX revert wrong previous fix and make weighted scorer test deterministic
MAINT skip unstable (arpack dependent) estimators in check_pipeline_consistency on Win32
Merge pull request #4295 from ragv/sigmoid_decision_fn
FIX & TST at least min_samples to be considered a core sample
DOC whats new for DBSCAN optims
DOC fix broken image link in feature selection
Merge pull request #4402 from amueller/ompcv_fix
Merge pull request #4388 from Barmaley-exe/svr-doc-n-test
Merge pull request #4407 from agramfort/calibration_nan_fix
Merge pull request #4370 from amueller/0.17_deprecated
Merge pull request #4416 from vortex-ape/kernel_precomputed
Merge pull request #4377 from vortex-ape/intercept_scaling
Merge pull request #4322 from amueller/kneighbors_include_self_fixes
Merge pull request #4368 from xuewei4d/deprecate_estimator_params
Merge pull request #4409 from bryandeng/immutable_defaults
Merge pull request #4422 from trevorstephens/y_numeric_for_np180
DOC whats_new entry for stabler StandardScaler
Merge pull request #4432 from ragv/travis_ignore_docstring
FIX test for older versions of numpy
Merge pull request #4427 from amueller/lda_lsqr_predict
Merge pull request #4439 from bryandeng/dataset_docs
Merge pull request #4350 from vortex-ape/build_issue
Merge pull request #4189 from amueller/sparse_decision_function
DOC fix link to Birch in whatsnew.rst
Merge pull request #4423 from vortex-ape/agg_clustering
DOC typo
DOC add warning for partial pip upgrades
FIX respect astype(array, dtype, copy=True) on old numpy
TST add missing test for astype backport
STYLE PEP8
Merge pull request #4542 from ogrisel/fix-doc-partial-pip-upgrades
FIX make shuffle / resample pass-through indexing utilities
MAINT remove debug nslookup call
STYLE PEP8
FIX make shuffle / resample pass-through indexing utilities
MAINT do not ctags index non-source artifacts
STYLE whitespace around operator in sklearn.utils.fixes
Merge pull request #4584 from sdegryze/fit-predict-for-pipeline
Merge pull request #4590 from jfraj/fix_bug4559
Merge pull request #4606 from ibayer/remove_order_check
Merge pull request #4608 from jfraj/test_pickle_bunch
Merge pull request #4604 from jnothman/faster_lshforest
Merge pull request #4610 from sseg/ex-wikipedia-py3
Merge pull request #4613 from bnaul/graph_lasso_tests
Merge pull request #4371 from saketkc/fix_setup
Merge pull request #4550 from amueller/common_test_refactoring
Merge pull request #4650 from lesteve/fix-4641
FIX remove deprecation warning
Merge pull request #4654 from amueller/fix_partial_dependency_percentiles
FIX #4597: LassoLarsCV on with readonly folds
Merge pull request #4683 from yanlend/patch-1
Merge pull request #4725 from jseabold/avoid-nan-equals-comp
DOC add comment for np.nan equality test in clone
FIX Gram OMP check_finite on old scipy
Merge pull request #4705 from amueller/cross_val_predict_input_types
Merge pull request #4697 from amueller/remove_warning_test_pandas_slicing
Merge pull request #4730 from amueller/ovr_classifier_sparse_coef
MAINT use the new container-based travis workers
Merge pull request #4741 from amueller/multioutput_warning
Merge pull request #4748 from TomDLT/check_maxiter
Merge pull request #4754 from TomDLT/astype_fix
FIX fix test_lars_path_readonly_data failure under Windows
FIX ResourceWarning in twenty newsgroups loader
Merge pull request #4362 from rvraghav93/make_PyFuncDistance_picklable
STYLE removed unused import, fixed comment indent and spacing
Merge pull request #4751 from amueller/dict_learning_n_jobs_bug
Merge pull request #4765 from alexeygrigorev/master
FIX typo in RobustScaler example
DOC various fixes in whats_new.rst
Merge pull request #4770 from TomDLT/logistic
FIX pass random_state to inner call to randomized_svd
STYLE cosmetic fixes in sklearn.mixture.gmm
MAINT remove referrence to HMM in common tests
Merge pull request #4872 from ogrisel/remove-hmm-ref-in-tests
MAINT make sure we do not download large datasets on appveyor
Merge pull request #4928 from rvraghav93/pls_deprecated
Merge pull request #4927 from jnothman/hiding_predict_proba
MAINT ignore Python import state in ctags
MAINT make it possible to use plain HTTP to download pre-built numpy & scipy from rackspace container
Merge pull request #4983 from ogrisel/appveyor-http-rackspace
Merge pull request #4908 from TomDLT/rcv1
Merge pull request #4984 from TomDLT/rcv1-subset
DOC updated whatsnew for joblib 0.9.0b2
Merge pull request #4961 from mrphilroth/issue4959
FIX seed the initialization of NMF
Merge pull request #5020 from ogrisel/fix-nmf-rng-seeding
Merge pull request #5017 from lesteve/fix-coveralls-badge
Merge pull request #5021 from lesteve/make-copy-joblib-sh-python3-compatible
Merge pull request #5005 from amueller/drop_old_doc_links
Merge pull request #4967 from stephen-hoover/threshold-boosting-samme_proba
Merge pull request #5019 from glouppe/voting-get_params
Merge pull request #4957 from dotsdl/issue-4614
Merge pull request #5113 from fzalkow/master
Merge pull request #5133 from arthurmensch/lasso_perf_checks
Merge pull request #5081 from amueller/transformers_consistent_n_samples
Merge pull request #5098 from olologin/OneClassSvm_sparse_test
DOC explain fork related issues in FAQ + stopgap for Python 3.4
Merge pull request #5167 from NoonienSoong/clustering_example
Merge pull request #5161 from beepee14/sparse_prediction_check
DOC add what's new entry for cross_val_predict fix
PEP8 in cross_validation module + tests
MAINT bump joblib to 0.9.0b4 to use forkserver for Py3 & POSIX
DOC update FAQ on multiprocessing forkserver
MAINT enable multiprocessing + kmeans test on Python 3.4
DOC whats_new entry for forkserver
FIX deprecation message for 1d data
Merge pull request #5228 from jmschrei/gb_apply
MAINT re-cythonize sklearn/tree/*.pyx
MAINT PEP8 / slightly faster tests in decomp LDA
Merge pull request #5249 from MechCoder/init_centroids_bug
OPTIM make (Online) LDA reuse a joblib.Parallel instance
Merge pull request #5206 from acganesh/numpy_scipy_ver_check
Merge pull request #5245 from ogrisel/lda-acronym-deprecation
Merge pull request #5280 from ogrisel/pr-4924-rebased
Merge pull request #5293 from amueller/pipeline_X1d_inverse
MAINT use inspect.signature for introspection
MAINT build and test with Python 3.5 on appveyor
MAINT update numpy / scipy wheels used by appveyor
Merge pull request #5303 from larsmans/faster-lda
MAINT configure the appveyor's cache for pip
Merge pull request #5302 from ogrisel/python-3.5-appveyor
ENH better error message for estimators with ensure_min_* checks
FIX consistency of memory layout for linear CD solver
FIX ensure contiguous Gram matrix in graph lasso
Merge pull request #5337 from ogrisel/fix-coordinate-descent-memory-layout
ENH utility to have distinct dataset .pkl filenames
FIX separate filenames for RCV1
FIX separate filenames for species distributions
FIX separate filenames for covtype
FIX separate filenames for 20 newsgroups
FIX separate filenames for california housing
FIX separate filenames for Olivetti faces
DOC what's new entry for dataset fetchers
Merge pull request #5356 from rvraghav93/ridge_appveyor_failure
Merge pull request #5386 from jmschrei/contrib
Merge pull request #5384 from map222/patch-1
Merge pull request #5362 from MechCoder/lasso_fix
Merge pull request #5234 from vighneshbirodkar/mcd_fix
Merge pull request #5378 from amueller/gridsearch_docs_unsupervised
Merge pull request #4478 from amueller/fix_randomized_svd_transpose
ENH use explicit decimal keyword & PEP8
DOC whats_new entry for randomized_svd heuristic
Merge pull request #5395 from amueller/some_test_deprecation_warnings
Merge pull request #5399 from lesteve/update-joblib-to-0.9
Merge pull request #5413 from amueller/cleanup_tests
Merge pull request #5411 from lesteve/fix-plot-tomography-l1-reconstruction-example
MAINT Release 0.17b1
Omer Katz (1):
Nest the for loops because they don't need to run if the condition is not true.
Pauli Virtanen (1):
TST: fix undefined behavior in test
Peter Fischer (1):
Typos in comments corrected
Peter Prettenhofer (5):
fix issue 4447 : min_weight_leaf not properly passed to PresortBestSplitter
add regression test case for issue 4447
cosmit
add huber to non-uniform sample weight toy test case
Merge pull request #4448 from pprett/fix-gbrt-min-leaf-weight
Phil Roth (9):
Correctly match penalty and loss with arguments
Add a space to an error message
allowing for different cluster stds
adding multiple cluster_std to make_blob test
adding assert statement to check the std
whitespace
Adding a new kmeans example.
Adding kmeans example to the narrative documentation
Adding a reference to the example in the narrative documentation
Preston Parry (1):
DOC Grammar fix in preprocessing
Raghav R V (36):
ENH Scale the sum_of_confidences to (-0.5, 0.5)
COSMIT remove trailing whitespaces
MAINT remove the deprecated n_iterations param
MAINT Remove the deprecated n_iterations from StratifiedShuffleSplit
MAINT docstring --> comments to prevent nose from using doc in verbose mode
MAINT use yield for cleaner output in verbose mode
DOC/TST docstring --> comment
FIX Include PyFuncDistance attributes while pickling.
TST Add test to make sure BallTree remains picklable even with callable metric
TST make sure DistMetric is picklable with both predefined as well as custom metric
MAINT merge _check_cv into check_cv as indices argument is removed in 0.17
REF Correct the reference link for additive chi squared sampler.
MAINT PLS Remove deprecated coefs attribute
FIX import column_or_1d from utils.validation directly
MAINT Remove support for the deprecated sequence of sequences
MAINT return 'unknown' for seq. of seqs in type_of_target
MAINT deprecate seq of seq in metrics fully with tests
FIX bailout w/ValueError if multilabel-seq or other unsupported types
MAINT/ENH inline the logic for _is_sequence_of_sequences
FIX make sure empty target vectors are returned as binary
FIX 2D binary array-like must be considered as unknown
TST Explicitly add one empty label
TST/MAINT reintroduce mixed types test sans seq. of seq.
MAINT is_label_indicator_matrix --> is_multilabel
FIX revert generation of seq of seq in make_multilabel_classification
MAINT remove unused import warnings
FIX type_of_target will now raise ValueError on ml-seq; [[1, 2]] is mc-mo;
TST (in metrics) change error message for seq. of seqs
DOC revert the documentation changes
MAINT Use check_is_fitted
FIX/REVERT undo removal of ignore_warnings to avoid a few warnings in tests
MAINT Remove ignore_warnings from test_matthews_corrcoef_nan
DOC Make cv documentation consistent across our codebase
FIX Use float64 to avoid spurious errors
FIX dtypes to conform to the stricter type cast rules of numpy 1.10
FIX Move validation from helper to main function
Rob Zinkov (1):
Adding sparse support for decision function
Robert Layton (1):
ENH precomputed is now a valid metric for 'brute'
Rohan Ramanath (1):
fixes scikit-learn/scikit-learn#5329
Saket Choudhary (2):
MAINT: Remove reimport
FIX: Ensure dependencies installed
Sam Zhang (1):
BUG: get_param in FeatureUnion disjoint for shallow/deep params
Sebastian Raschka (9):
ensemble classifier 1st commit
votingclassifier
added _check_proba method
property trick to raise AttributeError if voting is "hard"
inline check_proba
silhouette_score docstring update
changed clf to regr in decision tree regression examples
upd randomforest docstring
note about variance reduction
Sebastian Saeger (10):
Added a verbose flag to the GMM class.
Fixed typos.
Fixed doctests.
Fixed pep8 warnings.
Added tests for the verbose flag of GMM.
Fixed pep8 and removed unused variable.
Adapted the verbose flag of DPGMM. Boolean flag is deprecated now.
Added tests for the adapted verbose flag of DPGMM.
Removed the deprecation and added a test for a boolean flag value.
Cleaned the code. pep8, pyflakes and added comments.
Shivan Sornarajah (1):
Changed SVM to use gamma='auto' by default. Deprecated gamma=0.0.
SimonPL (2):
Rewrite of the documentation for LDA/QDA
Small edits to LDA/QDA documentation
Skipper Seabold (2):
TST: Regression test for nan parameters cloning
ENH: Avoid equality comparison for nans.
Sonny Hu (2):
fix RidgeCV when cv != None
add sample_weight into LinearRegression
Stephen Hoover (3):
DOC Fix doc for NMF `init` default
BUG Use epsilon threshold in `_samme_proba` and `_boost_real`
TST Increase samples for classifier tests
Steven De Gryze (3):
ENH implement fit_predict on pipelines
verifies fit_predict is available on last step of pipeline
better docstring one-liner and use of gettattr in test
Steven Seguin (4):
add py3 xrange to bicluster example
change to explicitly import function
Change xrange to range and remove patching import
Change urlopen and iteritems functions for Python3 compat
Theodore Vasiloudis (1):
[docs] [trivial] Fix small error in cross-validation docs.
Thomas Unterthiner (26):
ENH Added RobustScaler
Check if parameters are fitted
Fix documentation errors.
Remove warn_if_not_float
Remove 'copy' parameter form RobustScaler functions
Fix documentation in test case
Export robust_scale and RobustScaler
Improved documentation of robust scaler
Fixed decprecation warning
Changed outdated test
Fixed removal of copy-parameter
Removed superfluous testcase for robust scaler
Remove interquartile_scale parameter
Added robust scaling example
Fix robust_scaling test
Fixed documentation oversight.
robust_scaling example
Improved example
Fix documentation errors
More documentation fixes
Remove redundant code
DOC fix contributor name
ENH add MaxAbsScaler
ENH add minmax_scale
Add missing scalers to classes.rst
DOC Improve wording in minmax_scale documentation
Tiago Freitas Pereira (1):
Added the regularization term in the method _fit_transform
Tian Wang (1):
add a new int array to store indices
Tim Head (14):
Mention oob_score in narrative documentation of RandomForest*
Mention oob_score in the narrative docs for Bagging*
Added warm_start to bagging
Fix random_state to make test reproducible
Test for unchanged n_estimator and indentation fix
Test oob_score for warm_start'ed bagging classifiers
Removed overly deep indentation
oob_score will only be calculated if warm_start=False
Keep estimators_samples_ and estimators_features_ across warmstarts
Better assert_raises use, API compliance, better ordering
Ensure dict_learning_online always has at least one batch
Check fit() returns self for all estimators
Example to demonstrate use of tree.apply() method
Explicitly mention RandomTreesEmbedding in the text
Timothy Hopper (1):
Go to easy _open_ issues page
Tom DLT (1):
FIX temporary fix for sparse ridge with intercept fitting
Tom Dupré la Tour (1):
Merge pull request #5360 from TomDLT/sparseridge
TomDLT (17):
ENH improve check_array
ENH change func_grad_hess into grad_hess
ENH check that unfitted estimators raise ValueError
DOC update whats_new
ENH use astype to avoid unnecessary copy
FIX check parameter in LogisticRegression
ENH improve parameter check in LogisticRegression
FIX use random_state in LogisticRegression
DOC fix rendering of code-block
add fetch_rcv1
ENH implement LYRL2004 train/test split of rcv1
Reorder target_names in rcv1
ENH add sag solver in LogisticRegression and Ridge
ENH remove some warnings in test suit
FIX array_equal for numpy < 1.8.1
ENH refactor NMF and add CD solver
DOC update docstring and warning
Valentin Stolbunov (1):
Added handling of sample weights in logistic.py
Vighnesh Birodkar (7):
MAINT deprecate 1d input arrays for all estimators
added predict_equal_labels test and changed kmeans single
moved condition out of the loop
Added check for max_iter and test
Added explanatory comment
Fix warnings during tests
removed reshaping in fast_mcd and replaced by check_array
Vinayak Mehta (13):
added docstrings for restrict
Raising an error when intercept_scaling is zero
Specified original intercept_scaling value
Added docstrings for kernel='precomputed'
Raising an error when n_clusters <= 0
used special.fdtrc instead of fprob
Moved optimize test to a new file
Modified docs
Removed array2d from developer docs
Deprecated load_lfw_pairs and load_lfw_people
Changed import statement and checked for ImportError
MAINT Deprecate LDA/QDA in favor of expanded names
Fixed LDA typo in doc
Vincent (1):
Added random seed for facial recognition example and updated the docstring
Vincent Michel (1):
Fix RFE - Add the possibility to use feature_importances rather than coef_ when existing, see #2121
Wei Xue (8):
Deprecate estimator_params in RFE and RFECV #4292
Correct warning messages and move warnings to fit method #4292
FIX coding style #4292
Friendly error on KNeighbors(n_neighbors<=0 )
Raise ValueError when log_multivariate_normal_density has non positive definite covariance
DOC & COSMIT deprecate estimator_params in RFE and RFE_CV in docstring
ENH optimize rfecv by eliminating the for loop
Add regression test for the number of subsets of features
Yaroslav Halchenko (22):
LOOK INTO TODOS
initial changes for python3 pkgs
Merge tag '0.17b1' into releases
Merge branch 'releases' into dfsg
Merge branch 'dfsg' into debian
changelog
dropped CPed patches and refreshed one
requires joblib >= 0.9.2
debian/patches - deb_disable_google_anals to disable google tracking
BF: use systemwide underscore.js
BF: remove minimized jquery.js
Merge branch 'dfsg' into debian
slight tune up for license file -- BSD-2
- deb_nopaypal and deb_no_online_jquery
RF: use pandas rules approach to shuffle things into -lib etc
adjusted changelog, closes statements, point to debian branch for Vcs-Git
boost debian/compat, fix .installs etc
Remove minimized js files
Merge branch 'dfsg' into debian
also use non-minimized versions for jquery scripts
added maphilight copyright, updated years for the others
added overrides for the maphilight source
Yucheng Low (1):
Increase length of array indexing type in ArrayDataset
Yury Zhauniarovich (1):
Updated graph ranges
Zac Stewart (1):
Fix string literal concatenation whitespace
Zichen Wang (3):
added macro-average ROC to plot_roc.py
added macro-average ROC to plot_roc.py
rebased upstream/master with master
akitty (1):
Fix typo
banilo (1):
extend R^2 description
benjaminirving (1):
Adding checks for the input LDA prior
djipey (5):
Creation of the attribute LDA.explained_variance_ratio_, for the eigen solver.
Addition of a piece of docstring for the attribute
Correction of the name of a function.
Changing [ to (. Formatting consistency.
Modification of a line after @agramfort's suggestion.
edson duarte (1):
Fixed small typos in SVM examples.
gwulfs (2):
sklearn-theano
fix uri
jfraj (3):
Fixing bug in assert_raise_message #4559
Improving Bunch Class to ensure consistent attributes
Adding test to ensure bunch consistency after pickle
jnothman (18):
Merge pull request #4824 from amueller/testing_less_warnings
Merge pull request #4828 from untom/maxabs_scaler
Merge pull request #4853 from amueller/fix_ica_pca_example
Merge pull request #4859 from Titan-C/sidebar
Merge pull request #4862 from jnothman/fixlink
Merge pull request #4860 from rvraghav93/additive_chi2_sampler_reflink
Merge pull request #4840 from jnothman/dynamicgrid
Merge pull request #4934 from thvasilo/patch-1
ENH O(1) stop-word lookup when list provided
Merge pull request #4919 from rvraghav93/label_binarizer
DOC n_thresholds may be < no. of unique scores
Merge pull request #5094 from pv/fix-inplace
Merge pull request #5071 from mth4saurabh/fix-issue-3594
Merge pull request #5063 from amueller/bagging_input_validation
Merge pull request #5186 from sdvillal/issue5165
Merge pull request #5174 from jnothman/dbscan_precomputed_sparse
DOC what's new: DBSCAN sparse precomputed
DOC tweaks to what's new
maheshakya (2):
DOC Modified math expressions to theoretical bounds.
FIX Written in Latex format.
martinosorb (5):
Implemented parallelised version of mean_shift and test function.
Change par system to joblib and n_jobs convention
Minor appearance changes
Trivial function name bug fixed
pep8 style
mbillinger (4):
Added Testcase for LDA-Covariance
Fixed shrinkage covariance for LDA
Explicit broadcasting
Updated whats_new.rst
mhg (25):
lars_path method complemented with nonnegative option.
lars_path method complemented with nonnegative option.
Merge branch 'nonnegative-lars' of https://github.com/michigraber/scikit-learn into nonnegative-lars
"nonnegative" -> "positive" + cleanup
lars_path method complemented with nonnegative option.
"nonnegative" -> "positive" + cleanup
Merge branch 'nonnegative-lars' of https://github.com/michigraber/scikit-learn into nonnegative-lars
cleanup
todos for pull request
docstrings added + fit_intercept reintroduced.
cleanup
positivity constrained can now also be used for the CV estimators.
doctest fix.
positive option for Lars propagated to doc.
agramfort code review input
test lars_path positive constraint.
merge with upstream/master
fit_intercept docstring comment removed from lars_path method
positive option passed on to LarsCV estimator
tests of positive option for estimator classes
tests of positive option for estimator classes condensed.
pep-8 fix
tests for comparison of results for Lasso and LarsLasso added.
docstrings updated with considerations regarding small alphas when positive=True.
comparison test of Lasso and LassoLars under positive restriction refactored and commented
pianomania (3):
add a example for FeatureHasher
add an example for FeatureHasher
try to mollify travis
rasbt (1):
small typo fix "radius" in rbf kernel
santi (1):
ENH exposing extra parameters in t-sne
saurabh.bansod (1):
Fixes #3594.
scls19fr (2):
Update confusion_matrix docstring
Update confusion_matrix docstring
sinhrks (2):
DOC: Updated set_params doc
CLN: Added missing __all__
sseg (1):
Merge pull request #1 from scikit-learn/master
tokoroten (5):
randomforest decrease allocated memory
refactoring to simplify
refactoring to DRY.
change valiable name to meaningful.
autopep8 E501
trevorstephens (9):
fix LarsCV and LassoLarsCV fails for numpy 1.8.0
add gplearn to related projects
Pretty decision trees
update export_graphviz docs
unused params in gridsearchcv
add class_weight to PA cls, remove from PA reg
add sample_weight to RidgeClassifier
fixes #4846
deprecate fit params in qda and lda
unknown (5):
compare residual_threshold directly with 0.0
if there are no inliers, raise ValueError
fixed newlines, comments
Fixed test to check for capital No
changed format string in no inliers exception to be 2.6 compatible
vstolbunov (2):
Updated logistic regression tests with sag solver
Fixed syntax and combined two test functions
wangz10 (2):
fixed duplicated part in plot_roc.py
addressed jnothman's minor points.
zhai_pro (2):
FIX KMeans with fortran-aligned data.
TST: add a test for fortran-aligned data be used in KMeans.
Óscar Nájera (8):
sidebar hide on css
delete js of sidebar
A responsive behavior
Recover similar to old style
toggle shift to left larger vertical space
toggle contraction character
remove css for container-index that compensates old js behavior
css for container-index that compensates position of main content on index.rst and documentation.rst in small screens
-----------------------------------------------------------------------
No new revisions were added by this update.
--
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/scikit-learn.git
More information about the debian-science-commits
mailing list