bacbe9611d
Merge branch 'hcluster2'
...
Conflicts:
scikits/learn/datasets/samples_generator.py
scikits/learn/grid_search.py
2011-04-01 12:27:16 -04:00
e1ecda69b0
ENH: better doc and tests for unbalanced svm's
2011-03-31 12:01:57 +02:00
4699607491
DOC: some documentation for naive_bayes module.
2011-03-30 14:46:45 +02:00
ec96240c71
DOC: improve PLS docs and example
2011-03-25 00:27:18 +01:00
dded226da9
DOC: clearer doc for BallTree.
2011-03-23 15:55:40 +01:00
bf4268d3df
SGD: documentation for sample weights and class weights.
...
Some polishing of the SGD documentation.
2011-03-21 20:11:45 +01:00
3e9a383af6
Merge branch 'master' into sgdsampleweight
2011-03-21 08:04:44 +01:00
637c70038c
renamed load_* to fetch_* when network connection is potentially involved
2011-03-19 18:39:45 +01:00
f2fcd5ca5f
FIX: the datasets doctest fixture could never skip the tests when required
2011-03-19 17:21:35 +01:00
bc16f01e24
Additional tests for sample weights.
...
Fixed some typos in doc strings.
Changed signature of coef_ for SGDClassifier; now adheres to svm convention that coef_ is shape [1, n_features] in the case of binary classification.
2011-03-15 20:50:44 +01:00
2422ed9ef3
DOC: minor spellings and formatting (trailing spaces, consistent spacing etc)
2011-03-12 14:53:14 -05:00
d94119c7fe
DOC: Minor fixes to documentation
2011-03-09 22:58:53 +01:00
769155d9d1
resolve conflict
2011-03-09 15:10:57 +01:00
4d31f558b2
Merge branch 'master' of github.com:scikit-learn/scikit-learn
...
Conflicts:
doc/modules/classes.rst
scikits/learn/datasets/__init__.py
scikits/learn/datasets/base.py
2011-03-09 14:18:27 +01:00
27acf9100d
Merge branch 'master' into 20newsgroups-dataset
2011-03-03 18:50:58 +01:00
40f7ba34bd
ENH: Fixed seed for shuffling in SGD
...
Non reproducible random shuffling introduces too much noise in
cross-validation that makes parameter tuning hard.
2011-03-02 15:40:31 +01:00
09940a99a6
API: SVMs: eps -> tol
...
Change the eps argument to 'tol' and expose it in base classes.
The reasonning behind calling this argument tol rather than eps is that
it is a bound on the tolerance of the optimization, and it does not
relate to the machine precision (eps).
2011-03-01 17:51:50 +01:00
317df23fd8
Merge branch 'master' into 20newsgroups-dataset
2011-02-28 18:38:59 +01:00
8d8e02076d
Add metrics to the doc.
2011-02-28 18:20:11 +01:00
a8e76cc133
typo
2011-02-28 15:32:22 +01:00
a084f38160
Changelog for 0.7
2011-02-28 15:12:32 +01:00
1f710cf17f
improvements to the datasets documentation
2011-02-28 11:34:24 +01:00
36eb56d657
some more work on the datasets documentation
2011-02-28 09:56:35 +01:00
69c7ed5323
merge changes from LFW branch
2011-02-27 21:01:18 +01:00
adc953323c
merge lfw-dataset to 20newsgroups-dataset
2011-02-27 12:17:22 +01:00
123c0f8c11
better name: rename class_names to target_names for consistency
2011-02-27 12:15:50 +01:00
5b806ea408
better name: rename load_files to load_filenames
2011-02-27 12:08:35 +01:00
5c7ea36411
remove function autodoc section that breaks sphinx
2011-02-27 11:52:10 +01:00
62a5aae761
Merge branch 'master' into lfw-dataset
2011-02-25 02:05:37 +01:00
bef60efabc
documentation for the LFW dataset loaders
2011-02-25 02:04:37 +01:00
80673ed07d
Rename strategy --> algorithm in Neighbors*.
2011-02-24 20:03:16 +01:00
340761d8da
FIX: typo.
2011-02-23 10:53:24 +01:00
4a6e16fee5
Rename inplace --> brute_inplace
2011-02-23 10:53:24 +01:00
34cbb6556d
Faster Neighbors* in high dimensional spaces.
...
Add keyword mode={'auto', 'btree', 'brute', 'inplace'} to Neighbors.
This implements the possibility of using a brute-force algorithm when
the ambient space becomes too big. It also implements a mode='auto'
that uses a simple heuristic to use the best method.
2011-02-23 10:53:24 +01:00
c5aa155621
Merge branch 'hcluster' into hcluster2 that matches master
...
- pipeline and GridSearchCV have been cleaned
- feat agglo inherits from TransformerMixin
- s / n_comp / n_components in hierarchical.py
Conflicts:
scikits/learn/feature_selection/univariate_selection.py
scikits/learn/pipeline.py
scikits/learn/tests/test_pipeline.py
2011-02-21 21:28:34 -05:00
05386908b6
Merge branch 'pls' of https://github.com/fabianp/scikit-learn
2011-02-21 16:22:42 +01:00
39b3213d7d
DOC: set up barebones documentation for PLS.
2011-02-21 15:42:42 +01:00
ecd7190f09
Neighbors refactoring.
...
Rename NeighborsBarycenter --> NeighborsRegressor, Neighbors -->
NeighborsClassifier.
Also added parameter mode for NeighborRegressor (and updated examples
and doc describing it).
2011-02-17 10:37:34 +01:00
16c2d78654
Add reference for GCV.
2011-02-14 18:08:37 +09:00
0a8457bd0a
Minor fixes in RidgeCV.
2011-02-14 17:50:25 +09:00
1696b72ce7
Add RandomizedPCA to RST docs.
2011-02-11 13:58:19 +01:00
d5b445c28f
Doctest fixes.
...
The ellipsis thing seems to be mandatory for this to pass also on 32
bit systems.
2011-02-04 12:01:36 +01:00
5116c3d19a
DOC: svm.rst refactoring.
...
Also some other improvements to the doc and added example for SVR.
This closes issue #58 . Thanks to ShimaShima for the report.
2011-02-02 06:41:44 +01:00
da0807ccbb
adding example to rst doc
2011-01-31 22:05:48 -05:00
5e891f440d
Add documentation on matrices used for clustering.
2011-01-27 23:17:36 +09:00
99e870610f
Add documentation for the RandomizedPCA class
2011-01-27 00:59:51 +01:00
5f49bdfd24
Merge branch 'master' into asaf
2011-01-25 14:30:17 -05:00
de06c23ba7
in hierarchical : s/adjacency_matrix/connectivity, s/k/n_clusters
2011-01-25 13:06:06 -05:00
3d5dd2ece0
DOC: Add the logistic regression to linear models doc
2011-01-24 18:35:05 +01:00
bb5c923871
DOC: update RST doc for crossval with indices
2011-01-18 17:52:48 -05:00
Alexandre Gramfort