[scikit-learn] annotated tag 0.2-beta created (now 0bc5e88)
Andreas Tille
tille at debian.org
Wed Dec 28 13:10:53 UTC 2016
This is an automated email from the git hooks/post-receive script.
tille pushed a change to annotated tag 0.2-beta
in repository scikit-learn.
at 0bc5e88 (tag)
tagging 0e1faafec9871df73e875a0aadfcb67ec578c0e5 (commit)
replaces 0.1
tagged by Yaroslav Halchenko
on Thu Mar 25 08:45:54 2010 -0400
- Log -----------------------------------------------------------------
Tagging 0.2 beta release
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iEYEABECAAYFAkurWwIACgkQjRFFY3XAJMjtWgCgjfp4naHqStpbh6zQjcytQ8kj
nO4AoL8OYQK4da6fQvsxx01jGYkgWIKk
=l7MN
-----END PGP SIGNATURE-----
Alexandre Gramfort (28):
first attempt to handle standardization of data in SVM
bootstrapping lasso implementation with coordinate descent
adding myself to authors list
adding missing init file
bug fix in Lasso with coordinate descent
renaming lasso.py to cd.py
renaming in cd.py
adding support for elastic-net with coodinate descent
adding test for ElasticNet
adding implementation of ridge regression + test
adding paths for Lasso and ElasticNet + test
addind Lasso + ElasticNet example
cleaning Lasso + ElasticNet example
draft of documentation for GLM
minor fix of imports and plot
add util method for Leave One Out crossvalidation
add split method in LOO
adding missing svn keywords
cosmit on bayes
cosmit on crossval
minor fix
- bug fix in stop criterion of Elastic-Net
revert commit linreg -> glm
moving GLM cython code to src folder
renaming cd.py to coordinate_descent.py
renaming cd.py to coordinate_descent.py in example
speed improvement in coordinate descent methods
fix doctest issue in crossval (issue #34)
Bertrand Thirion (9):
added unisvaraiet selection. untested
some progress on feature selection: Ftest OK
first trial with svm. Does not work yet
added the tests for univariate selection
a small fix in univariate selection
temptative commit--deletion of feature-select messed up things
did the svn move manually
added a test to compare with stat_model
updated cross-validation
Fabian Pedregosa (117):
Update README with Sourceforge download section.
Remove machine prefix from namespace.
Fix import paths that where broken in last commit.
Fix typo in numpydoc (sphinx extension).
Updated docs to reflect the new namespace structure.
Some disutils cleanup and bug fixing.
Update neighbors module.
Delete old neighbors module.
Fix typo in setup.py
Start of 0.2 development cycle.
Update setup.py to reflect neighbors new location.
Remove redundant imports of cost_function
Do not compile modules that depend on boost by default.
New faithful dataset.
Datasets refactoring.
Import dap.client only locally.
Remove redundant info from project's setup.py
Fix typo in README
Delete web page, we will use the doc as webpage.
New bindings for libsvm.
Neighbors module refactoring.
Plotting sample for svm.
Delete old examples.
Add support for latex equations in sphinx.
Update contribute section.
Add sourceforge download area to install info.
Preliminary rst doc for svm.
Fix: do not scale by default in SVM.
Tests should not be executable (or nose will complain)
Clean bayes code.
remove empty dir
Bayes classifier object.
Put fast_logdet into utils.
Bayes return objects
use npy_intp instead of int
svn 64 bit fixes
revert unintended changes
Type changes in libsvm bindings.
Remove empty directories
Implement copy of parameters from svm (python) object to libsvm.
Refactor plot_svm examples.
Fix cython extension to work on 64 bit
Enhacement: Performance improvements for libsvm_helper.c
Flat is better than nested
Add Olivier Grisel to the AUTHORS file.
Rename of SVM classes.
Remove unused imports in svm's benchmarks
Update examples with new API changes.
Remove redundant info in libsvm_helper.c
Update docstrings in svm module.
Remove unused imports from cython files in glm/
Refactoring of the glm module.
Refactoring of the bayes module.
Add and update documentation for linear regression
Implement new dataset.
Fix performance issues in bayesian ridge regression.
Update examples to reflect new dataset
Move Bayesian Ridge Regression to module linreg.
Move remaining code in bayes to linreg.
Remove empty directories
Remove empty directories
Remove empty directories (yes, another one)
Remove empty directory ...
Remove empty directories (the last one, I hope)
Flat is better than nested.
Remove empty directories
Refactor svm model.
Add tests for linreg.Ridge.
Add some doc for the svm module.
SVM refactoring
Remove nonexistant directories.
Welcome Vincent.
Documentation for the svm module.
Cosmetic fixes.
update Iris dataset
Fix memory leaks.
tests for svm
svm
Fix segfaults in the case of SVR, OneClass.
Some dosctring and cosmetic changes to module svm
cosmetic changes to svm module.
Add svm example.
Fix bug when target vector dtype != int.
Add neighbors example
neighbors squash
Dataset refactoring.
Cometic commit, svm module
Beautify the svn example (colors where ugly!)
Documentation update.
Add another svm example.
Put images in a directory in doc.
Documentation update.
More doc for the svm module.
Fix spelling errors in svm doc.
Mark Ann module as deprecated.
Fix memory leaks in the svm bindings.
Update benchmarks
Update Neighbors module.
Remove unused functions from svm module.
Add module glm to parent's setup.py
Fix failing tests.
Fix distutils issues.
Cosmetic changes
Fix some docstring failures in crossval
Fix some docstring failures in svm
Comment failing doctests.
Remove failing doctests.
Fix doctest in neighbors module.
Fix doctests.
Update feature_selection.
Add small script to run all the examples.
Add missing directories to distutils
Do not run examples from tests in module em.
remove iris as python package (it is now just a data directory)
Removed test for em/tests/test_densities.py
ooops, I deleted too much in last commit.
0.2-beta release.
Gael Varoquaux (4):
ENH: Implement the feature selection the simplest way.
DOC: Add an example of feature selection.
ENH/API: Rewrite the leave one out to use a generator.
DOC: Beautify the lasso coordinate example and documentation.
Jake VanderPlas (1):
test: added my name to AUTHORS
Olivier Grisel (16):
cosmit + module docstring
first test for coordinate descent lasso
make sample_generator a package
cosmit
add comment in test
cosmit + docstrings
docstring for enet_dual_gab
test that the L1 and ElasticNet prior actually induce sparsity
cosmit
cosmit + docstring
cosmit
work in progress on callback system during fitting (not used yet)
leveraging the new callback feature for GLM to implement early stopping based on duality gab
typo
typo
give default values for alpha and beta and check that duality gap never increases for early stopping check
Vincent Michel (17):
First commit of bayes methods
Add convergenve threshold
Doc and corrections for ridge_regression
add the log-likelihood to the ridge regression function.
Add samples_generator folder for simulated data
Move samples_generator to datasets
Fix linear and add test for ridge
bayesian regression with ARD priors
Fix ARD and add the log-likelihood
Fix import bug
Correct API and nose tests for Linear Regression (Noprior, Ridge and ARD)
Some docs and examples - move bayes.bayes to bayes.regression
Add nonlinear samples generator
Cosmit
Change __repr__ for classes
Cosmit : add some __repr__ for classifiers classes.
Add the Lemma of Matrix inversion
-----------------------------------------------------------------------
No new revisions were added by this update.
--
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/scikit-learn.git
More information about the debian-science-commits
mailing list