081a554698
FIX P/R/F metrics and scorers are now for binary problems only by default
...
Scorers for different average parameters have been added.
2014-12-09 13:56:51 +11:00
787e4a0cea
DOC extend documentation on sample generators
...
Also allow make_multilabel_classification to return underlying
distributions.
2014-08-25 19:36:58 +10:00
5b247f90e4
move around examples for better structure.
2014-07-17 09:35:30 +02:00
2eb1e98eef
MAINT ensure that examples figures are displayed in the correct order
2014-07-14 15:47:17 +02:00
1ebb2287e0
Added :func: reference
2014-03-08 09:34:01 +05:30
4b32c8fa99
Added :func: reference
2014-03-08 08:26:25 +05:30
cbc005fc54
Added reference to function
2014-03-07 11:37:28 +05:30
613cf8e42c
s/n_features/n_samples in doc/datasets/index.rst
2013-11-05 20:58:59 +01:00
4b658aadeb
DOC add link to fetch_covertype to covertype narrative docs
2013-08-04 17:12:55 +02:00
b0035bf9e8
Rename cv_scores(_) back to grid_scores(_) to keep the name free for a future refactoring
2013-07-28 17:10:14 +02:00
b218edc793
started biclustering documentation and examples
2013-07-25 19:19:17 +02:00
798f442654
FIX faulty import in 20news docs
2013-07-23 13:43:20 +02:00
a75949a22c
DOC 20news filtering with smaller set and MultinomialNB
...
BernoulliNB in its present form has a hard time showing the good
features, because it isn't really a linear model (XXX fix this).
A smaller test makes this easier to reproduce for users.
2013-07-22 13:53:02 +02:00
abddaa8c69
ENH Add filters on newsgroup text
...
It is easy to overfit on the 20newsgroups dataset, by letting
classifiers learn from metadata that commonly appears in newsgroup
texts, but would be useless for identifying topics outside of this set
of newsgroups in 1993.
For example, many classifiers will tell you that three of the most
informative features are "nntp", "posting", and "host", because the
NNTP-Posting-Host header appears with different frequency in different
groups.
The fetch_20newsgroups function now allows you to ask for any of the
following kinds of text to be removed:
- Newsgroup headers (which contain lots of NNTP metadata that can
identify the group)
- Signature blocks (which often contain multiple terms that uniquely
identify the person posting, which in turn identifies the group)
- Quote blocks (which contain people's e-mail addresses and large
amounts of text from another post in the same newsgroup)
The 20newsgroups classification example takes the "--filtered" flag,
which will remove all of these. This noticeably decreases the accuracy
of all classifiers, leaving room for a better method to improve the
accuracy.
2013-07-22 13:53:02 +02:00
bbf74a94a0
DOC/FIX twenty_newsgroups.rst should use TfidfVectorizer
...
Instead of Vectorizer. Fixes #2173 .
2013-07-22 10:24:59 +02:00
8861833ec9
DOC spellfixes
2013-06-29 16:18:23 +03:00
176c9cee08
DOC no more mention of the Bunch in the narrative docs
2013-06-18 00:19:42 +02:00
b0f1d46deb
DOC fetch_20newsgroups returns the text, not text files. see SO question: http://stackoverflow.com/questions/16615523/using-scikits-kmeans-to-cluster-ones-own-documents
2013-05-19 16:45:39 +02:00
0859653e09
FIX: use the mldata mock in docstring as well
2013-05-01 13:49:58 +02:00
67ece78da4
COSMIT various typofixes
...
As suggested by codespell <https://github.com/lucasdemarchi/codespell >
Signed-off-by: Stefano Lattarini <stefano.lattarini@gmail.com>
2013-04-11 21:22:14 +02:00
3379504b6b
PY3: fix urlopen in mldata and california housing loaders
2013-03-19 17:56:57 +01:00
16e595ab94
ENH move covtype loading to sklearn.datasets
2013-02-13 21:01:19 +01:00
85ec0fd1ae
P3K death to the print statement
2013-02-11 23:11:57 +01:00
d454c13f9f
FIX #1417 : move nosetests configuration parameter to setup.cfg
2012-12-01 10:56:12 +01:00
ae82bc7a60
rm the long-deprecated scikits.learn package
2012-07-23 13:56:13 +02:00
ea6bac35f4
FIX: load_cities doesn't exist anymore
2012-06-01 18:24:21 +02:00
ebaee5ae18
Added more documentation on the cities dataset
2012-06-01 18:24:20 +02:00
4e3f00833e
DOC minor fixes to rst and image paths
2012-05-06 23:28:31 +02:00
10928c3f10
typo: Viola-Jones
2012-04-13 20:32:46 +02:00
835af05624
Merge remote branch 'upstream/master' into gradient_boosting
...
Conflicts:
sklearn/tree/tree.py
2012-03-13 07:16:35 +01:00
de3991a5b5
updated documentation: plots for gradient_boosting, new sample generator
2012-03-12 22:19:20 +01:00
4f31269ff9
DOC: started section on text feature extraction
2012-03-04 18:56:03 +01:00
9c44867355
FIX dataset docs: changed suffixes in include to match rename.
2012-02-21 00:51:52 +01:00
d24337b93f
DOCs change extension from txt to inc, add inc as doctest extension to makefile
2012-02-18 17:52:13 +01:00
dc74f817a8
DOC Insert hidden toctree, mv "included" files from rst to txt
2012-02-17 22:33:53 +01:00
5f4ca06734
FIX: make LFW data shapes consistent with Olivetti faces
2012-02-11 13:58:33 +01:00
5729aa21c6
DOC fix linking to load_sample_images and load_sample_image in docs
2012-01-08 20:32:03 +01:00
a17452244d
Cosmit in docs
...
Examples should be in 'topic' directive.
2012-01-03 08:49:22 +01:00
87a85c189a
Merge pull request #417 from larsmans/multilabel
...
MRG : ENH multilabel learning in OneVsRestClassifier
2011-12-21 08:10:23 -08:00
3d2ea0d3e6
Use int16 for more compactness.
2011-12-21 11:37:33 +01:00
9970fe2aa8
Add the samples generator to the references
2011-12-21 11:15:53 +01:00
dbbdbe40e6
Better doc for the 20newsgroup dataset loader.
2011-12-21 10:17:03 +01:00
9784dfd62e
Address @ogrisel and @amueller's comments.
2011-12-20 20:01:20 +01:00
8ec25b86e1
FIX This should fix the doctests in the datasets dir. They take quite long, I think it's because of the svmlight loaders. So I didn't include them in the standard make target
2011-12-18 19:35:22 +01:00
fa3c9dcffc
COSMIT: prettify datasets docs
2011-12-18 10:47:58 +01:00
ad019b600d
DOC: Corrections and additions to the dataset docs. Also more detailed docstrings
...
for the functions loading the data.
2011-12-18 10:47:58 +01:00
0343627513
DOC Description of the basic dataset API
2011-12-18 10:47:58 +01:00
dd0ce5b064
Merge pull request #467 from Balu-Varanasi/pep8-compliant
...
pep8 compliant
2011-12-17 02:52:53 -08:00
36d0adb8c1
pep8 compliant
2011-12-17 00:48:40 +05:30
e9be1c8af0
DOC: Added random dataset plot to doc.
2011-12-16 00:31:37 +01:00
Joel Nothman