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
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
f763a34a03
MAINT use yield for cleaner output in verbose mode
2015-03-21 11:17:32 +05:30
cd2ee7e454
MAINT docstring --> comments to prevent nose from using doc in verbose mode
2015-03-21 11:16:49 +05:30
fe553b755a
ENH use specific warning class for RP
2014-11-18 14:55:56 +01:00
9a6287b0e1
MAINT Fix regression random projections work sparse matrices
2014-07-30 18:34:26 +02:00
9fb2dfb56e
MAINT use assert_warns
2014-07-30 18:22:37 +02:00
c997cdee9f
COSMIT pep8 + full stop police
...
metrics/tests/test_score_objects.py made stricter by replacing . with \.
in regular expressions.
2014-02-01 15:28:53 +01:00
a6439d84b8
MAINT: randn on float is deprecated
2013-08-07 22:54:38 +02:00
966e10c252
COSMIT PEP8
2012-12-22 16:48:17 +01:00
a0860b114d
FIX don't use assert_allclose. It is not supported in numpy 1.3
2012-12-21 15:10:13 +01:00
8931e67bb5
DOC improve + fix mistakes + TST added
2012-12-21 14:48:01 +01:00
616e2b5d64
ENH Rename Bernoulli random projection to sparse random projection
2012-12-21 14:48:01 +01:00
12c8d3be79
ENH Rename Bernoulli random projection to sparse random projection
2012-12-21 14:48:01 +01:00
980600b67e
ENH refactor matrix generation BaseRandomProjectiona and subclass
2012-12-21 14:48:00 +01:00
a6ed58dc4d
ENH flatten random_projection module + add sklearn.utils.random
2012-12-21 14:48:00 +01:00
f014ed52db
ENH optim sparse bernouilli matrix
2012-12-21 14:34:23 +01:00
9a2d76a56b
COSMIT
2012-12-21 14:34:22 +01:00
825303ce3a
FIX renaming bug
2012-12-21 14:34:22 +01:00
f77185ebe0
ENH cosmit naming consistency
2012-12-21 14:34:22 +01:00
0bb6207382
ENH add more tests
2012-12-21 14:34:22 +01:00
3b96b9100f
DOC: correct doc
2012-12-21 14:34:22 +01:00
10aade4857
ENH add warning when user ask n_components > n_features
2012-12-21 14:34:22 +01:00
2e29be34a6
ENH add more tests with wrong input
2012-12-21 14:34:22 +01:00
956f3d8648
ENH Add gaussian projeciton + refactor sparse random matrix to reuse code
2012-12-21 14:34:22 +01:00
5aef6b2104
ENH Strenghtens the input checking
2012-12-21 14:34:22 +01:00
84bc401cdc
COSMIT
2012-12-21 14:34:21 +01:00
a5ebd1dec5
ENH let the user decide density between 0 and 1
2012-12-21 14:34:21 +01:00
2ac9b61db8
Clean random_dot features (3)
2012-12-21 14:34:21 +01:00
961c27a6c9
Clean random_dot features (3)
2012-12-21 14:34:21 +01:00
add00fa8c9
Clean random_dot features
2012-12-21 14:34:21 +01:00
35c75bfa77
ENH Let the user decide the number of random projections
2012-12-21 14:34:21 +01:00
edafb48416
COSMIT use sklearn.utils.testing
2012-12-21 14:34:21 +01:00
510b4a00c0
missing docstring and s/hashing_dot/random_dot/g
2012-12-21 14:34:21 +01:00
3621693afd
make it possible to use a preallocated output array for hashing_dot
2012-12-21 14:34:21 +01:00
08cb357456
leave dense_output=False by default
2012-12-21 14:34:20 +01:00
cbb2e39625
fixed broken seeding of the hashing_dot function
2012-12-21 14:34:20 +01:00
8e742d584a
refactored test to check that both materialized and implicit RP behave the same
2012-12-21 14:34:20 +01:00
ba5ea28c91
implemented dense_output=False for hashing_dot
2012-12-21 14:34:20 +01:00
2b329b906a
first stab at CSR input for hashing dot projections
2012-12-21 14:34:20 +01:00
a850892b21
OPTIM: call murmurhash once + update test & example
2012-12-21 14:34:20 +01:00
82cc48ffa3
working on implicit random projections using a hashing function
2012-12-21 14:34:20 +01:00
2feb067d35
WIP: starting implementation implicit random matrix dot product
2012-12-21 14:34:20 +01:00
62b9d2b983
check JL bound domain
2012-12-21 14:34:20 +01:00
f2d738dd97
change API to enforce dense_output representation by default
2012-12-21 14:34:19 +01:00
f0ece82ec1
more curified estimator API
2012-12-21 14:34:19 +01:00
c4d5d65ff0
leave self.density to 'auto' to implement the curified estimator pattern
2012-12-21 14:34:19 +01:00
60a86cc248
FIX: raise ValueError when n_components is too large
2012-12-21 14:34:19 +01:00
ea4392058d
OPTIM: CPU and memory optim by using a binomial and reservoir sampling instead of direct uniform sampling in the n_features space
2012-12-21 14:34:18 +01:00
fc1e9a6cde
test embedding quality and bad inputs (100% line coverage)
2012-12-21 14:34:18 +01:00
Raghav R V