3a172f3335
FIX SGD loss example for new hinge loss.
2012-02-01 22:52:37 +01:00
9e036a663e
MISC: add randomized linear models to what's new
2012-02-01 09:01:54 +01:00
5e0f82b664
Cosmit
2012-01-31 18:13:28 +01:00
a882bedcc3
MISC: address @ogrisel and @mblondel's comments
2012-01-31 18:11:05 +01:00
db1a2122f3
DOC: example and docs for randomized-lasso
2012-01-31 18:10:43 +01:00
51cf74290c
DOC: better example for randomized lasso
...
Highlight the regimes in which randomized-lasso beats L1 and F-scores,
and vice versa.
2012-01-31 18:10:43 +01:00
2b08a0ddae
DOC: enrich example docstring.
2012-01-31 18:10:43 +01:00
afb6dea4ac
ENH : fix legeng of plot_randomized_lasso.py
2012-01-31 18:10:43 +01:00
41a48638ec
ENH : early stopping in lasso_stability_path (faster)
2012-01-31 18:10:43 +01:00
b0b3406379
DOC: nicer path
2012-01-31 18:10:43 +01:00
2b1df1fb19
ENH: grid in terms of alpha/alpha_max
...
And rework example
2012-01-31 18:10:43 +01:00
5e840acedd
DOC: try to improve randomized lasso example
2012-01-31 18:10:42 +01:00
66d7b26d1c
cosmit in doc + pep8
2012-01-31 18:09:08 +01:00
b275030bdb
ENH : adding sample_fraction to lasso_stability_path + add to doc
2012-01-31 18:07:06 +01:00
19331b5ae3
DOC : adding doc for Randomized sparse linear models + fix test
2012-01-31 18:07:06 +01:00
f3423f8276
ENH : creating lasso_stability_path func + adding tests on randomized_l1
2012-01-31 18:06:14 +01:00
9c7deea99a
ENH : prettify plot_randomized_lasso.py
2012-01-31 18:06:14 +01:00
7ba93da27a
ENH : better plot_randomized_lasso.py with score path
2012-01-31 18:06:14 +01:00
6522b82b84
Jacknife in randomized_lasso
2012-01-31 18:01:02 +01:00
f7ca7b3a69
Better randomized_lasso
2012-01-31 18:01:02 +01:00
11aaf9816f
ENH: randomized_lasso example: multiple alpha
2012-01-31 18:01:02 +01:00
b0c119ca52
ENH: Add randomized lasso
...
Feature selection using Lasso (Meinshausen and Bulmhan 2008)
2012-01-31 18:01:02 +01:00
96d6060cd7
pep8
2011-12-23 10:12:26 -08:00
3f330f2967
STY : pep8
2011-12-20 15:34:17 +01:00
4c8aec3a6f
typo: accross => across
2011-12-20 09:49:45 +01:00
63331849ea
DOC: Fixed links to modules in the example gallery
2011-12-19 16:07:01 +01:00
976c16ecce
ENH : avoid global seeding in plot_polynomial_interpolation.py
2011-12-19 11:43:24 +01:00
63887c2b18
Style update in documentation.
2011-11-21 12:31:59 +01:00
87cf1b32dd
fix: sgd module clone issue w/ rho parameter
2011-11-15 21:06:32 +01:00
6e07284f3f
FIX example logisitic regression.
2011-09-14 12:53:35 +02:00
78dd8b5626
batch trailing spaces cleanup
2011-09-06 23:12:26 +02:00
4d1dcd2e65
DOC: scikits.learn -> sklearn
2011-09-03 09:10:40 +02:00
ddf4b72109
Move project directory from scikits.learn to sklearn
2011-09-02 12:06:57 +02:00
68f27e3790
Revert "Move project directory from scikits.learn to sklearn"
...
This reverts commit fd0d3b879d
2011-09-02 12:03:18 +02:00
fd0d3b879d
Move project directory from scikits.learn to sklearn
2011-09-02 11:38:24 +02:00
f8ef06eea0
FIX some broken examples.
2011-09-01 11:20:05 +02:00
43e3229a86
DOC: clarify doc on BIC/AIC
2011-08-10 16:30:31 -04:00
22da70185d
DOC: cosmetic changes to lars-bic doc and examples
2011-08-10 19:33:43 +02:00
f599fab517
ENH + DOC: addressing @GaelVaroquaux's comments
2011-08-10 16:12:41 +02:00
e4f4b907e7
ENH: addressing @ogrisel's comments on PR 298
2011-08-10 16:12:41 +02:00
899dfcda41
ENH : new LassoLarsIC estimator
2011-08-10 16:12:41 +02:00
a1d8259f4a
Merge branch 'master' into omp
...
Conflicts:
doc/modules/linear_model.rst
2011-08-05 17:50:34 +03:00
2cd0ccf71e
updated samples generator according to @glouppe's refactoring
2011-08-04 16:50:47 +03:00
626061357f
DOC: add example comparing LassoCV and LassoLarsCV
2011-08-03 20:03:22 +02:00
e2844ceef2
don't use gram in example, useless slowdown
2011-08-03 17:06:35 +03:00
945efb2233
API changes, part I
2011-08-03 16:18:16 +03:00
286dd1454e
BUG+COSMIT: rewamp the lasso path examples
2011-08-03 08:07:15 +02:00
bc20c29450
transposed generator, updated tests
2011-07-31 18:28:57 +03:00
33c2693839
unified notation
2011-07-29 13:05:07 +03:00
28a52f00cd
cosmit in example
2011-07-29 13:01:11 +03:00
f32ca3b162
Merge branch 'master' into omp
2011-07-29 12:55:28 +03:00
4ce86d6e02
pep8
2011-07-29 01:34:56 +03:00
50b0d828dc
Factored out sparse samples generator
2011-07-29 01:28:00 +03:00
5a12763441
API : renaming LARS to Lars
2011-07-25 21:41:48 -04:00
00af49f313
Remove useless prints in example
2011-07-25 14:08:27 +03:00
a2316af197
Brought in OMP from the larger branch
2011-07-25 13:57:42 +03:00
98ede0d989
Typo
2011-07-23 14:17:14 +02:00
eebc2ec325
DOC: Improve the Bayesian regression docs
2011-07-23 14:17:06 +02:00
6a1e2b31f2
pep8
2011-07-21 09:49:19 -04:00
6cd32e0976
BUG: s/empty/zeros in plot_lasso_bic_aic.py
2011-07-21 09:48:37 -04:00
d3502b09ad
ENH : adding LASSO model selection example based on BIC and AIC
2011-07-21 08:58:13 -04:00
298259070e
Improve Logistic Regression sparsity example.
2011-07-10 01:58:56 +09:00
05f54c2cda
DOC: new example for Ridge + more rst docs
2011-07-08 11:27:50 +02:00
f2f0caef27
DOC: Tweak the cross-val lasso path example
2011-06-18 09:33:14 +02:00
e44915fda8
the l1 logreg example now works with l1_min_c
2011-04-26 09:32:37 +02:00
66483cf2c0
Use np.vander.
2011-03-30 14:58:38 +09:00
b7c9201f6b
Add polynomial interpolation example.
2011-03-24 16:04:40 +09:00
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
dfa75da35d
Sample weights for SGDClassifier.
...
Additional test cases for class weight and sample weight.
Example plot for sample weight.
2011-03-15 19:58:56 +01:00
21175e3629
MISC: Fix example to adjust to eps -> tol
2011-03-02 06:52:58 +01:00
8864e0f32b
Remove example plot_lar.py
...
First, because it is wrong, as it prints the path of the Lasso and not
of Least Angle Regression. And second, because it is the same as
plot_lasso_lars.
2010-12-14 19:19:24 +01:00
89377b979a
API : maxit replaced by max_iter everywhere
...
API : removing default computation of r2 coef in coordinate_descent (speed up +
can be accessed simply eg. lasso.score(X, y))
2010-12-11 23:14:55 -05:00
7b94a59a59
pep8 + N->n_samples and D->n_features
2010-12-11 19:13:58 -05:00
e5ce482da3
revert the plot style to the LARS paper conventions
2010-12-09 18:57:45 +01:00
d9c7d80213
switch to base 10 for the alpha logs in the Lasso CD path plot
2010-12-09 18:56:32 +01:00
81b911c12b
make MSE plot of LassoCV more readable by scaling the y axis
2010-12-09 09:48:50 +01:00
f88ca8c4bb
ENH: make the LassoLARS and LassoCD path examples easier to compare
2010-12-09 09:32:32 +01:00
157fc0ac4c
ENH: LassoCV / ElasticNetCV now uses all folds data + example
...
Store all the mse information on each fold and select alpha by best median
on MSE.
2010-12-09 09:06:25 +01:00
085544eb5e
Use original diabetes data as shipped by the R package lars.
...
This also gets rid of stupid customizations in example.
2010-12-03 16:22:03 +01:00
46a2d551a3
fixed imports in non-auto examples.
...
COSMIT import paths in plot examples.
Rephrased SGD outline in linear_model.rst
2010-12-03 10:28:08 +01:00
d102d2228d
moved sgd examples to examples/linear_model and added sgd prefix.
...
minor fixes in some examples.
updated sgd.rst
2010-12-01 16:59:52 +01:00
68bc58d029
glm --> linear_model rename holocaust.
...
I also moved cd_fast.pyx from its src subdirectory to linear_models,
to match the sgd module structure.
2010-11-25 16:27:30 +01:00
Andreas Mueller