e3afc0e8c9
MAINT move custom error/warning classes into sklearn.exceptions
...
ENH NonBLASDotWarning -> EfficiencyWarning; Improve error message
DOC Add exceptions module to modules/classes.rst
MAINT Move ConvergenceWarning, UndefinedMetricWarning et al into exceptions
MAINT Remove ChangedBehaviorWarning from base
DOC/FIX Improve DataConversionWarning's docstring
2015-10-19 22:35:35 +02:00
d77ea1220d
Merge pull request #4707 from amueller/k_means_init_mismatch
...
[MGR] Raise error when init shape doesn't match n_clusters in KMeans
2015-10-07 17:51:09 -04:00
ddc985292f
Added check for max_iter and test
2015-10-06 14:31:31 -04:00
fc1875627c
moved condition out of the loop
2015-10-06 14:30:07 -04:00
687af98f9d
added predict_equal_labels test and changed kmeans single
2015-10-06 14:30:06 -04:00
0c1afd7e05
[BUG] _init_centroids has an optional x_squared_norms parameter which is not exactly optional
2015-09-10 15:14:32 -04:00
08bf350d4b
Merge pull request #5199 from ogrisel/joblib-0.9.0b4
...
[MRG+1] MAINT bump joblib to 0.9.0b4 to use forkserver for Py3 & POSIX
2015-09-08 18:04:13 -04:00
1529a1e548
MAINT enable multiprocessing + kmeans test on Python 3.4
2015-09-04 11:47:34 +02:00
343f9b4dcb
ENH sparse precomputed distance matrix in DBSCAN
2015-09-01 11:31:54 +10:00
0bf75367c9
Merge pull request #5189 from GaelVaroquaux/pr_4779
...
Pr 4779: Implemented parallelised version of mean_shift and test function
2015-08-30 19:49:24 +01:00
55c32ef25c
Merge pull request #4779 from martinosorb/parallel-ms
...
[MRG+1] Implemented parallelised version of mean_shift and test function.
2015-08-30 16:28:27 +01:00
c4721f36d8
TST: No n_jobs=-1 in the test
...
Use n_jobs=2, to avoid blowing up on hugely parallel machines.
2015-08-30 16:00:57 +01:00
847e366a2e
FIX test with invalid input; simplify dbscan precomputed
2015-08-30 13:49:46 +01:00
a8626b36a6
TST/COSMIT remove nose call boilerplate
2015-05-28 14:54:01 -04:00
4c4b1f7067
Change par system to joblib and n_jobs convention
2015-05-28 12:23:13 +01:00
7041ed283d
Implemented parallelised version of mean_shift and test function.
2015-05-27 17:46:00 +01:00
89c1018c64
ENH improve check_array
...
ENH improve check_array to warn on dtype conversions
ENH make check_array accept several dtypes
ENH change validation with improved check_array
ENH change astype to avoid copy if possible
ENH remove warn_if_not_float
2015-05-19 16:13:42 +02:00
ec1bba7f52
FIX / TST raise error when init shape doesn't match n_clusters in KMeans, check for sensible errors.
2015-05-13 09:58:35 -04:00
af1b652080
FIX DBSCAN fit with precomputed matrix in edge cases
...
and add test.
2015-04-30 13:40:59 +02:00
d0302d629e
TST: add a test for fortran-aligned data be used in KMeans.
2015-04-07 01:43:12 +08:00
56546a7a32
Merge pull request #4423 from vortex-ape/agg_clustering
...
[MRG + 1] Raising an error when n_clusters <= 0 in AgglomerativeClustering
2015-03-31 10:15:43 +02:00
cd2ee7e454
MAINT docstring --> comments to prevent nose from using doc in verbose mode
2015-03-21 11:16:49 +05:30
ea5474bb77
Raising an error when n_clusters <= 0
...
Changed ValueError wording
2015-03-21 04:23:35 +05:30
cd4fef9a4a
Merge pull request #4322 from amueller/kneighbors_include_self_fixes
...
[MRG+2] Pass include_self=True to kneighbors_graph
2015-03-20 13:05:10 +01:00
6beacc3fb6
remove deprecated stuff from 0.17
2015-03-18 14:49:04 -04:00
2e99294a54
FIX & TST at least min_samples to be considered a core sample
2015-03-17 16:21:10 +01:00
46060feddc
Pass the appropriate include_self argument to kneighbors_graph everywhere.
2015-03-16 13:30:22 -04:00
15c9c0f01a
FIX/TST boundary cases in dbscan ( closes #4073 )
2015-03-05 23:31:54 +01:00
a2087d8eee
Do not shuffle in DBSCAN (warn if `random_state` is used).
...
This makes little difference, and original DBSCAN did not shuffle.
Warn if `random_state` is used.
As is `random_state` encourages users to experiment with different
randomization, as you would do with k-means. But in contrast to
k-means, the output of DBSCAN is deterministic except for cluster
enumeration and "rare" cases, where a point is on the border of
two clusters at the same time. As this affects single points only,
the measureable performance difference will be close to zero.
Also, incorporate fix for minpts including the query point.
2015-03-05 23:31:54 +01:00
f1b8283c95
catch some warnings, be less verbose in testing.
2015-03-02 14:40:35 -05:00
e07d420bb9
FIX: Fix n_components in AgglomerativeClustering and friends
2015-02-19 22:19:59 +05:30
5c0c5c672c
MAINT use absolute imports in tests
...
as per the guideline in:
http://scikit-learn.org/stable/developers/#coding-guidelines
2015-02-10 13:40:31 +01:00
a9b3965b0a
Better error messages in MeanShift, slightly more robust to bad binning.
2015-02-07 12:59:04 +01:00
e3f94351f2
TST Silence some tests.
2015-01-29 17:24:41 -05:00
500f032c4e
Merge pull request #4148 from amueller/clean_up_after_tests
...
[MRG+1] remove tempfiles that we create in testing
2015-01-24 15:22:03 +01:00
fc51c062ce
FIX agglomerative clustering w/ connectivity constraints & precomputed distances
2015-01-24 22:37:06 +11:00
0a4c12f58a
TST remove tempfiles that we create in svmlight tests and agglomerative clustering tests
2015-01-23 11:36:46 -05:00
494b8e5743
TST add test for sparse matrix handling in clustering
...
Make sparsity check check everything.
don't test everything. That would be nice but is out of scope :-/
catch special case of no core samples in DBSCAN
add nonregression test for sparse dbscan with no core samples.
2015-01-16 12:03:20 -05:00
4fe8d93835
MAINT Remove the _check_fitted test.
2015-01-12 00:33:40 +05:30
40f44ffd23
MANIT: pep8ize to an extent cluster module
2015-01-11 13:22:28 +05:30
39c34178d8
DOC/COSMIT clean up according to review
2014-12-25 14:22:41 +11:00
684cfb3151
FIX DBSCAN.fit_predict now also supports sample_weight
2014-12-25 14:22:24 +11:00
11569e3f13
ENH support weighted samples in DBSCAN
2014-12-25 14:22:24 +11:00
8a240f110d
ENH sparse support in DBSCAN
2014-12-25 14:22:24 +11:00
2182262bcc
Merge pull request #3985 from MechCoder/return_distances
...
[MRG] ENH: Return distances option for linkage_trees
2014-12-22 22:17:36 +01:00
bf6275428a
TST: Test ground truth in linkage trees
2014-12-21 23:06:44 +01:00
860f77356f
FIX: Fix bug in computing full tree when n_clusters is large
2014-12-20 02:46:17 +01:00
acb924f310
Merge pull request #3952 from MechCoder/callable_connectivity
...
[MRG+1] ENH: Allow connectivity to be a callable in AgglomerativeClutering et.al
2014-12-20 00:02:35 +01:00
acc25a47fd
Merge return_distance tests of both linkage and ward tree
2014-12-19 18:20:57 +01:00
324780355a
ENH: Return distances option for linkage_trees
2014-12-19 18:00:11 +01:00
Raghav R V