b460ffb8fe
ENH: added PresortBestSplitter
2013-09-21 18:26:28 +02:00
2eddb1a29f
BUG fix whitespace in error messages
...
Partial fix for #2380 .
Signed-off-by: Lars Buitinck <L.J.Buitinck@uva.nl>
2013-08-26 17:02:52 +02:00
fd3a816afb
renamed supported_loss to _SUPPORTED_LOSS (constants)
...
add test for deprecated warning
2013-08-24 19:04:11 +03:00
b952cdb052
GBRT checks if ``loss`` is in self.supported_loss
2013-08-24 18:08:04 +03:00
716fbb062c
much better input validation, test that warning is raised on (n_samples, 1) y
2013-07-28 16:08:16 +02:00
3edb5b1b72
COMPAT: string formatting syntax in Py 2.6
2013-07-27 16:46:44 +02:00
c38caacb3d
Refactored verbose output in GBRT - output much more nice
...
Added tests for verbose output in GBRT
cosmit: use str.format instead of %
updated whats new with GBRT changes
2013-07-25 14:24:57 +02:00
bf063b145e
rearanged some code
2013-07-24 11:17:33 +02:00
04e8d2befd
narrative documentation for oob improvement.
2013-07-24 11:17:32 +02:00
9c58296e48
fix: oob_improvement_ with trailing _
2013-07-24 11:17:32 +02:00
34b3f03b6c
add OOB improvement and set oob_score deprecated
2013-07-24 11:17:31 +02:00
e488704849
COSMIT pep8
2013-07-23 15:17:32 +02:00
80dba53607
WIP: check inputs for gbrt
2013-07-22 12:36:55 +02:00
2c3861fc1f
WIP: Make trees and forests support string labels
2013-07-22 12:20:02 +02:00
f03095a978
WIP: use np.ascontiguousarray instead of array2d
2013-07-22 12:20:02 +02:00
bbdccca354
WIP: enforce contiguous arrays to optimize construction
2013-07-22 12:20:02 +02:00
f84e463ae9
WIP: remove irrelevant and unstable doctests
2013-07-22 12:20:00 +02:00
52b193a239
WIP: PEP8 a few things
2013-07-22 12:20:00 +02:00
247172e9f8
WIP: remove BreimanSplitter
2013-07-22 12:20:00 +02:00
3f540dafa6
WIP: small optimizations
2013-07-22 12:19:58 +02:00
8f18d96e5f
WIP: Added BreimanSplitter
2013-07-22 12:19:58 +02:00
314ffce266
FIX: doctests
2013-07-22 12:19:58 +02:00
4742806106
gbrt now works with new DecisionTree implementation
2013-07-22 12:19:57 +02:00
dacfd8bd5d
DOC: Replaced all BSD style licenses with "BSD 3 clause"
...
Replaced all BSD style licenses with "BSD 3 clause"
Not checked yet!
Removed duplicate "3 clause, 3 clause"
Removed trailing period if exists
Fixed some missed licences, still about 50 to do, but those can be automated
Think I got the last of them.
Apparently me and sed have different ideas of regex.
Found a few more
2013-04-30 08:34:46 +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
eae18a3784
PY3 fix metaclasses. See #1829 .
2013-04-04 01:07:20 +06:00
495b6c3f1b
DOC: GBRT max_features
2013-02-25 15:15:26 +01:00
c41ccbb25c
ENH: Defer parameter checking of trees
2013-02-25 15:13:16 +01:00
42c7496c0c
Merge branch 'master' into gbrt-deviance-fix
2013-02-17 19:52:21 +01:00
5c3dc9a029
ENH: Update feature_importances in GBRT
2013-02-07 18:27:52 +01:00
f7f5424464
Remove compute_importances parameter
2013-02-05 16:19:32 +01:00
11b126f3bd
Merge branch 'master' into gbrt-deviance-fix
2013-01-26 18:37:06 +01:00
9ddb9ca2d2
correct docstring
2013-01-26 18:27:21 +01:00
e369c4339d
COSMIT remove unused imports, pep8
2013-01-26 18:26:20 +01:00
caeb299415
raise ValueError if division through zero in LogOddsEstimator
2013-01-26 17:25:11 +01:00
83026a8a0a
fix: deviance computation in BinomialDeviance was wrong (ignored cases where y == 0) - thanks to ChrisBeaumont for reporting this issue
2013-01-26 12:19:42 +01:00
07c56d7cd2
COSMIT rm deprecated stuff
...
One more deprecated feature to go in FastICA; not sure how to handle
that one.
2013-01-22 19:45:27 +01:00
c0323d1d57
FIX remaining parameter issues in GradientBoosting
2013-01-15 22:35:10 +01:00
696dd6fb6b
FIX in tree and ensemble: don't overwrite random state in fit.
2013-01-15 22:35:09 +01:00
78fc091a2c
FIX: set min_samples_split=2 by default
2013-01-08 23:29:19 +01:00
754c2d9533
COSMIT pep8
2012-12-22 16:48:17 +01:00
2b56ee1ba3
set min_density when growing deep trees during gradient boosting
2012-12-14 20:45:45 +01:00
9c94b3770a
ENH add author
2012-12-11 09:35:12 +01:00
6c60fe214a
FIX feature importance computation + Enable smoke test of feature importances
2012-12-11 09:33:11 +01:00
f4f34d492d
PEP8
2012-12-11 09:33:11 +01:00
2c71362478
COSMIT
2012-12-11 09:33:11 +01:00
af50266a3c
ENH preserver contiguous property
...
The a[:,np.newaxis] operation does not preserve contiguousity, so we shift to
reshape which preserve this one. For an example see below:
In [1]: import numpy as np
In [2]: np.__version__
Out[2]: '1.6.2'
In [3]: a = np.array([1,2,3])
In [4]: a.flags
Out[4]:
C_CONTIGUOUS : True
F_CONTIGUOUS : True
OWNDATA : True
WRITEABLE : True
ALIGNED : True
UPDATEIFCOPY : False
In [5]: a[:,np.newaxis].flags
Out[5]:
C_CONTIGUOUS : False
F_CONTIGUOUS : False
OWNDATA : False
WRITEABLE : True
ALIGNED : True
UPDATEIFCOPY : False
In [6]: a.reshape(-1,1).flags
Out[6]:
C_CONTIGUOUS : True
F_CONTIGUOUS : False
OWNDATA : False
WRITEABLE : True
ALIGNED : True
UPDATEIFCOPY : False
2012-12-11 09:33:11 +01:00
44ba8edec4
COSMIT
2012-12-11 09:33:11 +01:00
e8d21a798f
ENH Move parameter checking to fit
2012-12-11 09:33:11 +01:00
77064a2822
DOC explain parameter estimators_ (2)
2012-12-11 09:33:11 +01:00
Gilles Louppe