Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Comparing changes

Choose two branches to see what's changed or to start a new pull request. If you need to, you can also compare across forks.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also compare across forks.
base fork: zaxtax/scikit-learn
...
head fork: zaxtax/scikit-learn
Checking mergeability… Don't worry, you can still create the pull request.
This comparison is big! We're only showing the most recent 250 commits
Commits on Nov 04, 2013
@MechCoder MechCoder DOC Minor doc changes 0546d31
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency 501230d
@MechCoder MechCoder Removed unnecessary attributes 23331a9
@pprett pprett sparsify test: add multi-class test too c9b6d14
@oddskool oddskool completed sparsity sections; added benchmark 1bdb7b2
@oddskool oddskool cosmit link e86a580
@oddskool oddskool cosmit double backticks 6241ac7
@oddskool oddskool precisions from @ogrisel 4052f64
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency f39c759
@ogrisel ogrisel Merge pull request #2571 from edran/readme
Added contributing subsection to README
5f8f7a0
Commits on Nov 05, 2013
@ogrisel ogrisel Only run make test when coverage is disabled 84f86c2
@ogrisel ogrisel Merge pull request #2574 from ogrisel/travis-speed
[WIP] Only run make test when coverage is disabled
d6e5462
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency 7a6d653
@oddskool oddskool @ogrisel precision on sparse + elasticnet tradeoff bc54b7b
@oddskool oddskool link to benchmark in github c6bae4c
@ankit-maverick ankit-maverick Returning an array mask using np.ma.getmaskarray 28997f6
@oddskool oddskool fix for np.count_nonzero not present in Numpy < 1.6 c7c6999
@GaelVaroquaux GaelVaroquaux MAINT: make gen_rst more robust (minor) 7b3e7c7
@agramfort agramfort s/n_features/n_samples in doc/datasets/index.rst 613cf8e
Commits on Nov 06, 2013
@ankit-maverick ankit-maverick Adding a test for verifying the shape of imputed matrices 6f8186f
@ankit-maverick ankit-maverick Using assert_equal routine from nose 067ace1
@oddskool oddskool whats_new.rst 3fff76d
@agramfort agramfort Merge pull request #2576 from ankit-maverick/issue2560
Fixing issue 2560 : Imputer bug with median and dense input
f719d09
@mblondel mblondel Test correctness of average_precision_score. d0cdcde
@mblondel mblondel Merge branch 'master' of github.com:scikit-learn/scikit-learn 63f7dd1
@mblondel mblondel Remove some warnings. b7f98ac
Commits on Nov 07, 2013
@mblondel mblondel Add docstrings to _auc and _average_precision. 6d09f05
@oddskool oddskool s/speed/latency/g 179d758
@oddskool oddskool added section on model complexity 4ae976b
@pprett pprett fix: Huber loss function in gradient_boosting fails if negative_gradi…
…ent is not called before __call__; now computes on-demand.
c5ce49b
@pprett pprett add huber loss bug fix to whats new 623dfef
Commits on Nov 08, 2013
@glouppe glouppe Merge pull request #2566 from Manoj-Kumar-S/constant-output
Constant output dummy classifier.
bf033bf
@glouppe glouppe What's new: DummyClassifier now supports constant output a150413
@oddskool oddskool now compares linear vs svr vs rf 87a01e1
Commits on Nov 09, 2013
@mblondel mblondel BUG: ValueError was assigned as local variable! 8143e40
@mblondel mblondel COSMIT: no need for parentheses. c0e686b
@mblondel mblondel More robust unit tests for fast_dot. 9162dd5
@mblondel mblondel Remove warnings in ProbabilisticPCA tests. be2894a
@mblondel mblondel One less warning in QDA tests. d6e722a
Commits on Nov 10, 2013
@mblondel mblondel Simplify fast_dot. 112befd
@amueller amueller DOC explanatory sentence for svm grid search example dd15031
Commits on Nov 11, 2013
@agramfort agramfort Merge pull request #2583 from amueller/doc_rbf_parameters
MRG DOC explanatory sentence for svm grid search example
24a06c9
@mblondel mblondel Remove print statement. 5944e1e
@mblondel mblondel Add lobpcg in possible solvers. 281017c
@mblondel mblondel Fix fragile symmetry check. 31d747b
@mblondel mblondel Fix fragile test.
Failed on two of my machines.
2a23fdc
@mblondel mblondel Fix typo in test. 215ca71
@larsmans larsmans DOC remove confusing comment from TruncatedSVD
The comment was supposed to suggest that there might be a faster
way to do fit_transform than the matrix-vector multiplication
that it currently performs, but the suggested alternative actually
has higher complexity.
317e97a
Commits on Nov 12, 2013
@jakevdp jakevdp ENH: add PolynomialFeatures preprocessor 568c43d
@ogrisel ogrisel FIX: errors in kernel parameters for OneClassSVM a8089b0
@jakevdp jakevdp address PR2585 comments d071ef1
@jakevdp jakevdp import pylab -> import matplotlib.pyplot eb22538
Commits on Nov 13, 2013
@jakevdp jakevdp TST: fix PolynomialFeatures test error 217834b
@jakevdp jakevdp TST: fix doctests & pep8 2983ab9
@jakevdp jakevdp DOC: adjust notes in PolynomialFeatures ecdedd5
Commits on Nov 14, 2013
@oddskool oddskool added example and fullfilled section on influence of model complexity dd84e10
@oddskool oddskool removed CV code to tune hyperparameters (better for CI) 0371831
@oddskool oddskool mentionned training throughput dc763a3
@oddskool oddskool added ElasticNet to the narrative doc 8e2d6a2
@oddskool oddskool more descriptive legend for ElasticNet 4427c7a
@jnothman jnothman FIX limit warnings for recall_score, precision_score, f1_score,
fbeta_score
ae4d0ac
Commits on Nov 15, 2013
@jakevdp jakevdp reorder polynomial features 4c00042
@oddskool oddskool merge doc/whats_new.rst f08d2c6
@jakevdp jakevdp fix ordering & docs of PolynomialFeatures f372e12
Commits on Nov 16, 2013
@jakevdp jakevdp TST: remove stray print statement 6715860
Commits on Nov 17, 2013
@MechCoder MechCoder ENH: ElasticNetCV and LassoCV raise ValueError with multitarget outputs af8904c
Commits on Nov 18, 2013
@larsmans larsmans DOC: multiclass: make the warning more prominent
Too many newbies are asking questions involving a
OneVsRestClassifier(LinearSVC()) or similar construct
on the various forums.
e804575
@larsmans larsmans DOC improve feature selection docs e2b6cc7
@larsmans larsmans COSMIT remove useless "if False" in kernel approximations
Callable case handled by pairwise_kernels.
af43c88
@bdkearns bdkearns FIX regression in CountVectorizer handling of float min_df/max_df
Fixes #2595.
da4f2c6
@jnothman jnothman Merge branch 'master' of https://github.com/scikit-learn/scikit-learn
…into prf_derivative_warnings
21f8fd0
@jnothman jnothman COSMIT use assert_no_warnings where appropriate a23662d
Commits on Nov 19, 2013
@yarikoptic yarikoptic BF: use hasattr with providing attr name (Thanks to Timo Schulz) e5f1cec
Commits on Nov 20, 2013
@GaelVaroquaux GaelVaroquaux BUG: fix partial_fit of MiniBatchDictLearning
The inner state of the algorithm (essentially sufficient statistics
approximating the data already seen) were not kept, and as a result
partial_fit was strongly sub-optimal
e5bed51
@GaelVaroquaux GaelVaroquaux DOC: update whats_new 86954e2
Commits on Nov 21, 2013
@larsmans larsmans DOC explain ARPACK algorithm in TruncatedSVD docstring
Fixes #2572.
b3aff11
Commits on Nov 22, 2013
@jakevdp jakevdp DOC: typo in kernel formulas cfa7775
@jakevdp jakevdp Merge pull request #2608 from jakevdp/master
DOC: typo in kernel formulas
7ca3906
Commits on Nov 23, 2013
@jakevdp jakevdp Merge pull request #2585 from jakevdp/polynomial
[MRG] ENH: add PolynomialFeatures preprocessor
f63e5fd
Commits on Nov 25, 2013
@NelleV NelleV FIX verbosity of the mds
Now, the verbose option doesn't need to be strictly equal to 2 to print out
the stress at each iteration.
96cd50a
@craigdsthompson craigdsthompson MAINT: unreachable code removed from BernoulliNB
The deleted code was unreachable. The previous call to check_arrays
ensures that X and y have consistent first dimensions.
36d68b3
@larsmans larsmans BUG SVMlight loader should check whether n_features is big enough
scipy.sparse won't complain when it gets a dimensionality that is
too large, causing crashes in estimators.

Fixes #1630.
119cc24
@dengemann dengemann MAINT: use stable blas getter API 642cbb2
@larsmans larsmans BUG lower space complexity of estimate_bandwidth to linear
Previously quadratic, see #2613.
669e2b4
Commits on Nov 26, 2013
@zyv zyv BUG: typo fixes in sklearn.mixture.gmm dea7bbb
@arjoly arjoly Merge pull request #2614 from zyv/patch-1
BUG: typo fixes in sklearn.mixture.gmm
f642aee
@rgommers rgommers MAINT: fix broken links to numfocus.org on donations page. b5e197f
Commits on Nov 27, 2013
@GaelVaroquaux GaelVaroquaux Merge pull request #2615 from rgommers/fix-numfocus-link
MAINT: fix broken links to numfocus.org on donations page.
a7d19e3
@pprett pprett fix: pass warm_start to BaseSGDRegressor
fix: make Regressor test case extend CommonTest (move two clf specifc tests to clf test case)
c02dee5
@larsmans larsmans DOC be more explicit on mean-shift scalability
Fixes #2613.
e0cb201
@pprett pprett Merge pull request #2617 from pprett/sgd-regressor-fix
[MRG] pass warm_start to BaseSGDRegressor
80b96bc
Commits on Nov 28, 2013
@jnothman jnothman FIX stop forcing deprecation warnings for external packages 9e1edd1
@jnothman jnothman FIX use of str.format for Py2.6 61c3b6a
@jaquesgrobler jaquesgrobler speedup for docbuild using joblib 2373961
Commits on Nov 29, 2013
@jaquesgrobler jaquesgrobler change cache directory 293c020
@GaelVaroquaux GaelVaroquaux Merge pull request #2620 from jaquesgrobler/master
ENH: speedup for docbuild using joblib
d4fd210
@jaquesgrobler jaquesgrobler fix broken link for stable version from dev version feb4312
@dengemann dengemann FIX: unintended assignment to ValueError 4d01a77
@dengemann dengemann STY: tuple for Gael 5055836
@dengemann dengemann ENH: refactor warnings 1 d2f3a8a
@dengemann dengemann ENH: add assert_warn_message e1bdd99
@dengemann dengemann ENH: revem tests_metrics.py f9cd599
@dengemann dengemann ENH: more of this 3a6c034
@dengemann dengemann ENH: final fix: clear all regisitries 9b1207b
@dengemann dengemann FIX: PPCA related warnings a0f0a3d
@dengemann dengemann ENH: context manager + decorator 9fd2be4
@dengemann dengemann FIX: doc example a4e67d6
@dengemann dengemann ENH: address discussion 163e659
@dengemann dengemann ENH: address discussion 4c10c82
@dengemann dengemann rebase c300814
@dengemann dengemann cosmits e141a98
@dengemann dengemann FIX
FIX typo
686fc3a
@dengemann dengemann FIX typo 2 26a5dd7
@dengemann dengemann address discussion 39fb5c5
@dengemann dengemann remove import 0b62e57
@dengemann dengemann Improve doc d8ad680
@GaelVaroquaux GaelVaroquaux BUG: fix convergence check in OMP dc8a03e
@larsmans larsmans Merge pull request #2541 from dengemann/fix_dot
[MRG]: fix tests, centralize warnings and reset `__warningregistry__`
8b037ad
@dengemann dengemann py3k syntax 1eacd31
Commits on Nov 30, 2013
@agramfort agramfort Merge pull request #2624 from dengemann/fix_dot
FIX: py3k syntax in recently added doctest
391c913
Commits on Dec 01, 2013
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency 09d7e97
Commits on Dec 02, 2013
@dengemann dengemann ENH: color scheme + add PCA c025984
@dengemann dengemann COSMIT 93f2a9a
@dengemann dengemann address duscssion 2b1863b
@dengemann dengemann address duscssion 2 + expose modern API d990ad9
@dengemann dengemann ICA vs PCA barrier free + fixed 05188ef
@dengemann dengemann COSMIT efe5294
@arjoly arjoly TST add generic test for averaging 5821143
@arjoly arjoly ENH test that most metrics work with one sample 9e56092
@arjoly arjoly ENH auc averaging for multilabel-indicator format e90317d
@arjoly arjoly ENH add multilabel-indicator support for average_precision 5b9c369
@arjoly arjoly TST perform testing on average precision + simplify invariance test s…
…ystem
5ddea8c
@arjoly arjoly ENH add scorer support for multi-label threshold scorer 3df6e8a
@arjoly arjoly TST scorer works with multi-output decision function afc82c5
@arjoly arjoly flake8 7ee45a5
@arjoly arjoly TST more clean up test_metrics 205a422
@arjoly arjoly DOC narrative doc for averaging support of average_precision_score 348a342
@arjoly arjoly FIX decision function scorer multilabel e9b4fcf
@arjoly arjoly DOC narrative doc for roc_auc_score multilabel-indicator e05cb16
@arjoly arjoly DOC not all functions support multilabel-sequence format 873a58c
@arjoly arjoly DOC wording c53e977
@arjoly arjoly FIX doctests 87b2785
@arjoly arjoly DOC extend roc example fbb62f4
@arjoly arjoly DOC update example of precision-recall curve 8832bde
@arjoly arjoly TST properly raise ValueError... eb56b9e
@arjoly arjoly DOC nicer plot e778c36
@arjoly arjoly TST add roc_curve and precision_recall_curve on toydata 7bafbca
@arjoly arjoly DOC TST explain how to use common tests in test_metrics a9aa616
@arjoly arjoly Typo 59fd981
@arjoly arjoly TST clean copy paste mistake 85e1db0
@arjoly arjoly TST remove print 303a465
@arjoly arjoly TST full coverage for _average_binary_score da1b56a
@arjoly arjoly DOC TST typo 3d0211c
@arjoly arjoly WIP TST more sample invariance test 19bf47e
@arjoly arjoly WIP TST adapt test for sample based metrics 1aa0f07
@arjoly arjoly TST finish to add invariance test for sample based metrics 5cf711a
@arjoly arjoly DOC typo 41ecb9b
@arjoly arjoly DOC wording and typo 3c4dc02
@arjoly arjoly TST add test for garbage averaging input string 909d45d
@arjoly arjoly TST typo 050f7ec
@arjoly arjoly DOC explain why roc auc score is useful in multilabel classification 49186bb
@arjoly arjoly typo 8c2c1c5
@dengemann dengemann Improve wording + COSMITS f3c09df
@dengemann dengemann Improve wording + COSMITS II 925aa1e
@dengemann dengemann manual subplots adjust
manual subplots adjust
2843ee1
@agramfort agramfort Merge pull request #2628 from dengemann/improve_ica_example
ENH: color scheme + add PCA
527d5f7
@ogrisel ogrisel Merge pull request #2460 from arjoly/auc-multilabel
[MRG] Multilabel-indicator roc auc and average precision
a82c8ac
@ogrisel ogrisel FIX: broken sparse matrix check under scipy 0.13.0 2e07f4e
@ogrisel ogrisel FIX: Python 3 dict keys cannot be concatenated with a list 04977db
@oddskool oddskool found a good elasticnet example 3e7f445
@arjoly arjoly TST more corner case tests 7a01a5c
@arjoly arjoly TST ensure that exceptions are properly raised on np 1.3 848eab3
@arjoly arjoly Merge pull request #2629 from arjoly/fix-auc
Fix average precision score test on numpy 1.3
06a1eaf
@oddskool oddskool fixed link & desc in WN.rst 7c9d618
Commits on Dec 03, 2013
@oddskool oddskool added model complexity in latency/throuhputs example plots 2723270
@oddskool oddskool added prediction time plot to Reuters example f4d32b2
@oddskool oddskool typos in narrative doc 3b7bc1f
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency d101ef4
@arjoly arjoly Update what's new 87343d7
@oddskool oddskool Joel's feedbacks 3b254d5
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency 6ec80b5
Commits on Dec 04, 2013
@pprett pprett fix: l1_ratio is incorrect; its (1.0 - rho); added test case to make …
…sure elastic net penalty with l1_ratio near to 1 and 0 matches L1 and L2 penalties.
30578da
@pprett pprett added change log entry 11d256c
@pprett pprett set l1_ratio to (1.0 - l1_ratio) 6a25717
@pprett pprett revert: reverted l1_ratio change - not SGDRegressor - sorry 3721988
@ogrisel ogrisel Merge pull request #2631 from pprett/fix-sgd-l1-ratio
[MRG] Fix sgd l1 ratio
d27cbac
@josericardo josericardo Fixing small typos in the docs. a1aa449
@arjoly arjoly Merge pull request #2633 from josericardo/FixingTypos
Just fixing small typos in the docs.
7abe96d
@ogrisel ogrisel ENH: more explicit traceback + message in case of common failure on s…
…parse input
699079e
Commits on Dec 05, 2013
@oddskool oddskool Merge remote-tracking branch 'upstream/master' into pred_latency 54b03b3
@oddskool oddskool added model re-shaping + compressed verbose § cdf70f4
@ogrisel ogrisel Merge pull request #2592 from jnothman/prf_derivative_warnings
FIX limit warnings for recall_score, precision_score, f1_score,
4674552
@oddskool oddskool emph. reshaping benefits on I/O a3a4b53
@jnothman jnothman ENH support multiclass targets of string objects de7bee4
@larsmans larsmans MAINT: use $(MAKE) for recursive make
For people who want to use custom make, e.g. GNU make on Solaris or BSD.
d025750
@larsmans larsmans DOC add sklearn.base to generated docs
Fixes #1449.
7643f44
@larsmans larsmans ENH use fast row_norms helper in preprocessing.normalize
Also makes memory usage independent of n_features.
c94ea6a
@larsmans larsmans ENH use fast row_norms in dictionary learning 21f0479
Commits on Dec 06, 2013
@GaelVaroquaux GaelVaroquaux Merge pull request #2488 from oddskool/pred_latency
[MRG] DOC Prediction Speed
66a5a4a
@larsmans larsmans BUG don't densify sparse matrix in BernoulliRBM.score_samples
Also documented the non-determinism of this method.
b0485c4
@larsmans larsmans ENH speed up progress reporting in RBM
Previously took up to 45% (!) of the time in training, not the 10% promised
in the docstring, for the digits example. Now down to at most 3%.

Pseudo-likelihood is now reported for the entire X, rather than batch-by-batch,
which will take more memory.
4706435
Commits on Dec 07, 2013
@jnothman jnothman TST Add benchmarking script for multilabel metrics fa5acb2
@larsmans larsmans ENH better error message when CountVectorizer prunes away all terms
Fixes #2596.
d02a15b
@larsmans larsmans TST disable non-doctests in Comp. Perf. docs
Btw., why is this in the modules/ directory?
be744a0
@larsmans larsmans DOC {min,max}_df look at document freq, not term freq dcb2834
@GaelVaroquaux GaelVaroquaux Merge pull request #2626 from jnothman/string_objects
ENH support multiclass targets of dtype=object strings
393a894
@jnothman jnothman DOC Add what's new entry for #2626 3aeae7b
@larsmans larsmans DOC fix BernoulliRBM._fit docstring
Return value is gone now. Thanks @jnothman.
8b6b770
@larsmans larsmans Merge pull request #2642 from larsmans/rbm-speedup
[MRG] faster convergence reporting in RBMs
f2b91dc
Commits on Dec 08, 2013
@larsmans larsmans ENH: faster cartesian product in make_classification and document com…
…plexity

This version
* is 4-6x faster
* takes less memory, and
* (on my box) is more likely to give a decent MemoryError instead of
  triggering the OOM killer, because it allocates in one go.

See #2534.
03352bf
@larsmans larsmans Revert "ENH: faster cartesian product in make_classification and docu…
…ment complexity"

This reverts commit 03352bf.

Apparently doesn't work with older NumPy.
4b85423
Commits on Dec 09, 2013
@trein trein Fix exception message when cloning a estimator that does not implemen…
…t 'get_params' method.
912d550
@agramfort agramfort Merge pull request #2647 from trein/patch-1
Fix exception message when cloning an estimator
baac91c
@jnothman jnothman COSMIT in response to feedback 84ec4f9
@arjoly arjoly Merge pull request #2643 from jnothman/bench_mutilabel_metrics
[MRG] Add benchmarking script for multilabel metrics
5f57f85
@jakevdp jakevdp DOC: PolynomialFeatures narrative doc 88bff86
@jakevdp jakevdp DOC: fix auto-links c66d3df
@jakevdp jakevdp DOC: fix Polynomial Regression docs 6cb1e3d
@jseabold jseabold ENH: Raise explicitly on non-unique vocab. 0bd0552
Commits on Dec 10, 2013
@jaquesgrobler jaquesgrobler remove EllipticEnvelop deprecation 79c1c38
@GaelVaroquaux GaelVaroquaux Merge pull request #2653 from jaquesgrobler/remove_ellipticenvelope_d…
…eprecation

remove EllipticEnvelop deprecation
b308a53
@arjoly arjoly Remove deprecated zero_one and zero_one_score a64d5aa
@ogrisel ogrisel Merge pull request #2649 from jakevdp/polynomial_doc
DOC: Polynomial regression narrative doc
34005e9
@arjoly arjoly flake8 c80116c
@GaelVaroquaux GaelVaroquaux Merge pull request #2654 from arjoly/metrics-0.15
[MRG] Remove deprecated zero_one and zero_one_score
ded381b
@jakevdp jakevdp DOC: fix test failure in linear model ffde690
@amueller amueller Website: add spotify testimonials 25ed3f5
@jaquesgrobler jaquesgrobler Merge pull request #2651 from amueller/spotify_testimonial
MRG Website: add spotify testimonials
833a1f3
Commits on Dec 11, 2013
@arjoly arjoly FIX silence numpy 1.8 warning for using non integer 20694a8
@arjoly arjoly Merge pull request #2656 from arjoly/np1.8-warnings
[MRG] FIX silence numpy 1.8 warning for using non integer
f86ed77
@jakevdp jakevdp SPD: reuse variable in GaussianNB for speedup dba9ade
Commits on Dec 12, 2013
@mblondel mblondel Merge pull request #2659 from jakevdp/gaussiannb_speedup
SPD: reuse variable in GaussianNB for speedup
aa8139b
@amueller amueller ENH make minibatch k-means use batches in prediction and computing la…
…bels
9671b3c
Commits on Dec 13, 2013
@agramfort agramfort FIX : explained_variance_ratio_ in RandomizedPCA 71d2392
Commits on Dec 14, 2013
@agramfort agramfort update tests 4ce2eb7
@agramfort agramfort better sparse support 9b37285
@agramfort agramfort pep8 08469a6
@agramfort agramfort update what's new 2b8d06b
@agramfort agramfort add test with sparse data 0553bf1
Commits on Dec 15, 2013
@agramfort agramfort address @ogrisel's comments fe1c284
@AndreaCatalucci AndreaCatalucci fixed bug in mds.py
this error is raised in OSX 10.9 with float64: use np.allclose, make code cleaner and less buggy
22322f1
@AndreaCatalucci AndreaCatalucci added not 8e58fcb
@GaelVaroquaux GaelVaroquaux Merge pull request #2667 from kowalski87/master
fixed bug in mds.py
9d92982
Commits on Dec 16, 2013
@ogrisel ogrisel Merge pull request #2664 from agramfort/fix_randomized_pca
FIX : explained_variance_ratio_ in RandomizedPCA
8dabff1
@jnothman jnothman Merge pull request #2660 from amueller/mbkmeans_compute_labels_fix_new
MRG make minibatch k-means use mini-batches in prediction
5b300e5
@GaelVaroquaux GaelVaroquaux DOC: add phimeca testimonial 05cce23
@agramfort agramfort Merge pull request #2669 from GaelVaroquaux/phimeca
DOC: add phimeca testimonial
aac7e67
@agramfort agramfort fix f_oneway with ints 44270dd
@agramfort agramfort pyflakes f31d6d7
@GaelVaroquaux GaelVaroquaux Merge pull request #2671 from agramfort/fix_f_oneway_int
fix f_oneway with ints
d4c5b18