2f099334f9
MAINT deprecate 1d input arrays for all estimators
...
Passing 1D arrays to check_array, without setting `ensure_2d` to false now
raises a deprecation warning before reshaping it. This will later throw an
error.
All Scaler classes also throw warnings when 1D arrays are passed.
All unit tests/doctests are modified to ensure that no 1D arrays are passed,
except in explicit 1D array tests where the warnings have been silenced.
Additional tests are also included which check for different 1D array cases.
2D array tests with one samples and one features are also added and where
they failed, `check_array` call has been modified to give a more useful error
message
2015-09-09 15:49:58 +02:00
664d78eb7c
MAINT Remove support for the deprecated sequence of sequences
...
MAINT Remove sequence of sequence support from datasets
MAINT Remove return_indicator param
MAINT Remove multilabel-seq test in OVR
MAINT Remove multilable-seq test in check_cv
MAINT Remove multilabel seq test in label_binarizer
TST type_of_target returns "unknown" for multilabel-sequence types
TST _check_targets should raise a ValueError
DOC show multilabel indicator as an example; remove return_indicator param
DOC use consistent lower case y for target
2015-08-11 22:26:44 +05:30
b09afb08be
DOC/MAINT final touches to FunctionTransformer
...
* copyedit doc
* give the module an underscore (we should do this to all private modules...)
* use public import path in test
* change all logs to log1p to prevent warning in test and because it's
genuinely useful for frequency data
* what's new
2015-08-03 22:12:19 +02:00
ba44e7bcc3
COMPAT: Makes test_function_transformer py2 compatible.
2015-08-03 11:45:16 -04:00
9d14f7352e
ENH: Renames CallableTransformer -> FunctionTransformer.
...
Makes `pass_y` an argument to FunctionTransformer to indicate that the
labels should be passed to the wrapped function.
2015-08-03 11:45:16 -04:00
cb0916c440
ENH: Adds CallableTransformer
...
CallableTransformer allows a user to convert a standard python callable
into a transformer for use in a Pipeline.
2015-08-03 11:45:16 -04:00
7fd2206f4f
Merge pull request #4960 from mkrump/lb_encoder
...
[MRG + 1] Don't allow unseen values for inverse transform
2015-07-28 22:22:56 +02:00
06f16e2634
Added test
2015-07-11 17:15:33 -05:00
26fbfe6f2e
ENH add minmax_scale
2015-07-01 23:27:35 +02:00
ab734b7a39
ENH add MaxAbsScaler
2015-06-11 16:08:17 +02:00
0b07536990
Merge pull request #4125 from untom/RobustScaler
...
[MRG + 2] ENH RobustScaler
2015-05-26 18:02:41 -04:00
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
d86a629ac7
Added robust scaling example
2015-05-16 11:15:32 +02:00
dd2fdbb18c
Removed superfluous testcase for robust scaler
2015-05-16 01:50:56 +02:00
e5fb9dd5a4
Changed outdated test
2015-05-16 00:50:25 +02:00
4ff47d9efa
Fix documentation in test case
2015-05-15 23:51:17 +02:00
007ae76cb3
TST: covering norm='max' branches of normalize()
2015-05-08 14:20:54 -04:00
a94301cad2
added an additional test case in test_label.test_label_binarize_with_class_order()
2015-04-02 18:25:52 -05:00
574ebfda85
Merge pull request #4436 from ogrisel/rebased-pr-3747
...
[MRG+2] FIX make StandardScaler & scale more numerically stable
2015-03-23 11:12:32 -05:00
b596dfcfef
FIX test for older versions of numpy
2015-03-23 16:41:18 +01:00
04078be7da
FIX make StandardScaler & scale more numerically stable
...
Detect common numerical stability problems in the standardization
routines and try simple mitigation strategies while alos issuing
a warning.
2015-03-23 14:32:33 +01:00
cd2ee7e454
MAINT docstring --> comments to prevent nose from using doc in verbose mode
2015-03-21 11:16:49 +05:30
6beacc3fb6
remove deprecated stuff from 0.17
2015-03-18 14:49:04 -04:00
ed47b5b744
Directly compute polynomial features.
...
Polynomial features are computed by iterating over all combinations
of features. For each combination of features, the product of the
columns indexed by the combination is computed.
The fit method is now a no-op, and the transform method works with any
number of features (regardless of what fit was called with).
2015-03-18 13:07:15 +11:00
22f718dcce
ENH Added RobustScaler
2015-01-19 21:13:31 +01:00
48e72b5559
TST Add test to check if ValueError is raised if input contains NaN
2015-01-16 19:13:31 +05:30
731b50bed6
TST Add test to check if warning is raised for integer dtype-ed inputs.
2015-01-16 19:12:13 +05:30
d71c52929a
Handle_unknown option to OneHotEncoder
...
OneHotEncoder fails with a error that is not helpful if a missing
categorical feature is present during transformation.
Add handle_unknown with an "error" and an "ignore" option.
2014-10-24 13:23:53 +02:00
cd7b43ccf9
TST make catch_warnings blocks more robust
2014-10-12 18:20:18 +02:00
52adb5cae7
ENH Add 'axis' argument to sparsefuncs.mean_variance_axis
2014-09-03 13:47:48 +02:00
cbcae0459c
FIX Support unseen labels LabelBinarizer and test
...
This fix restores the default behavior of 0.14.1 for backward
compat.
2014-07-29 16:21:04 +02:00
9f45e3730f
Modified sparse OvR to handle sparse target data
...
Defaulted label binarizer to set sparse_output=True when training ovr
classifiers, edited Label binarizer to allow for sparse binary column output
2014-07-18 12:10:50 +02:00
7768a3059b
ENH: enable y to only implement the array interface
...
For instance y could be a pandas' dataframe
2014-07-17 11:44:07 +02:00
9c1217ce5b
TYPO remove mutli (did you mean Muttley?)
2014-07-15 12:09:45 +02:00
b65e4c8d9d
FIX MinMaxScaler behavior on 1D inputs.
2014-07-13 18:35:21 +02:00
5358be563a
TST crank up preprocessing tests to 99% coverage
...
Fixes #3347 . (nosetests reports no missing lines, so 99% seems to
be a roundoff error, or a hint that 100% coverage is impossible :p)
2014-07-06 14:43:26 +02:00
2bfca14c49
ENH sparse matrix support in label binarization
2014-06-25 16:15:19 -04:00
60e8aeaf22
TST test interaction features outside doctest
2014-06-07 15:00:41 +02:00
28eb1ebe15
Assert or ignore all sequence of sequences deprecation warnings
2014-06-05 11:20:36 +02:00
9cfa1f38ad
TST stronger test for non-integers in MultiLabelBinarizer
2014-06-05 11:20:36 +02:00
5b4a3c8d41
TST Validate input MultiLabelBinarizer.inverse_transform
2014-06-05 11:20:36 +02:00
4dfde8bd54
FIX No set construction shorthand in Py2.6
2014-06-05 11:20:36 +02:00
bbc3e7e668
COSMIT use mlb for MultiLabelBinarizer instances
...
This class had been called LabelSetsBinarizer, for which lsb was appropriate
2014-06-05 11:20:36 +02:00
742505d870
FIX remove duplicates in MultiLabelBinarizer
2014-06-05 11:20:35 +02:00
18c399e20a
MAINT deprecate sequences of sequences support
...
* add warnings
* provide alternative binarizer, sklearn.preprocessing.MultiLabelBinarizer
* fix documentation
2014-06-05 11:20:35 +02:00
724b63a16c
Replacing the test .todense() methods with .toarray()
2014-05-20 21:44:34 -04:00
32c843b4b9
COSMIT: pep8, trailing spaces
2014-03-29 17:59:35 +01:00
14b435c136
Merge pull request #2938 from jnothman/clean_impute
...
[MRG+1] some clean-up in Imputer, particularly in calculation of sparse median
2014-03-11 19:18:05 -07:00
43da4ad830
ENH/TST remove unnecessary sorts and complete testing for sparse median
2014-03-04 00:16:49 +11:00
276e260486
ENH add sparse parameter to OneHotEncoder
2014-03-01 17:21:24 +01:00
Vighnesh Birodkar