* Add the new BayesianGaussianMixture class.
Add the test file for the BayesianGaussianMixture.
* Add the use of the cholesky decomposition of the precision matrix.
* Fix some bugs.
* Modification of GaussianMixture class.
The purpose here is to prepare the integration of BayesianGaussianMixture.
* Fix comments.
* Modification of the Docstring.
* Add license and author.
* Fix pb typo of eq 10.64 and 10.62.
* Correct VBGMM bugs.
* Fix full version.
* Fix the precision normalisation pb.
* Fix all cov_type algo for BayesianGaussianMixture.
* Optimisation of spherical and diag computation.
* Code simplification.
* Check the Gaussian Mixture tests are ok.
* Add test.
* Add new tests for BayesianGaussianMixture and GaussianMixture.
* Add the bayesian_gaussian_example and the doc.
* Fix comments.
* Fix review comments and add license and author.
* Fix test compare covar type.
* Fix reviews.
* Fix tests.
* Fix review comments.
* Correct reviews.
* Fix travis pb.
* Fix circleci pb.
* Fix review comments.
* Fix typo.
* Fix comments.
Add reg_covar and what's new.
* Fix comments.
* Fix comments.
* [ci skip] Correct legend.
Pipeline and FeatureUnion steps may now be set with set_params, and transformers may be replaced with None to effectively remove them.
Also test and improve ducktyping of Pipeline methods
* docs: fix broken and redirect links
see #7000
* docs: fix links
see #7000
* docs: merge with master
* docs: fix fnrs and tinyclues logo links
* docs: fix link reference in text
* docs: fix typo
* docs: added back in metaoptimize-qa paragraph
* docs: update language for defunct site
* docs: update stackexchange section
* docs: remove defunct site, move quora to top
* docs: remove defunct link and rearrange links
* Fix: Replace pylab with matplotlib.pyplot #6754
- one instance of 22 occurrences of pylab replaced with matplotlib.pyplot
- bench_glm.py now free of pylab references
- code executes properly
* Fix: Replace pylab with matplotlib.pyplot #6754
- one instance of 21 remaining occurrences of pylab replaced with
matplotlib.pyplot
- bench_glmnet.py now free of pylab references
- code does not execute for extraneous reason: ImportError: No module named
glmnet.elastic_net
* Fix: Replace pylab with matplotlib.pyplot #6754
- one instance of 19 occurrences of pylab replaced with matplotlib.pyplot
- bench_lasso.py now free of pylab references
- code executes properly
* Fix: Replace pylab with matplotlib.pyplot #6754
- one instance of 18 occurrences of pylab replaced with matplotlib.pyplot
- bench_plot_neighbors.py now free of pylab references
- code executes properly
* Fix: Replace pylab with matplotlib.pyplot #6754
- one instance of 17 occurrences of pylab replaced with matplotlib.pyplot
- bench_plot_omp_lars.py now free of pylab references
- code does not execute for extraneous reasons:
- File "bench_plot_omp_lars.py", line 111, in <module>
- ax = fig.add_subplot(1, 2, i)
- ValueError: num must be 1 <= num <= 2, not 0
- line 111 should probably be ax = fig.add_subplot(1, 2, i+1)
* Fix: Replace pylab with matplotlib.pyplot #6754
- bench_plot_parallel_pairwise.py now free of pylab references
- code executes properly
* Fix: Replace pylab with matplotlib.pyplot #6754
- bench_plot_ward.py now free of pylab references
- code executes properly
* Fix: Replace pylab with matplotlib.pyplot #6754
- bench_sgd_regression.py now free of pylab references
- code executes properly
* Fix: Replace pylab with matplotlib.pyplot #6754
- bench_tree.py now free of pylab references
- code executes properly
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_glm.py clean
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_glm.py clean of pl
- code does not execute for extraneous reasons
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_lasso.py clean of pl
- code executes properly
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_plot_neighbors.py clean of pl
- code executes properly
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_plot_omp_lars.py clean of pl
- code does not execute for extraneous reasons
* fix: Fix bug that prevented graphs from displaying
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_plot_parallel_pairwise.py clean of pl
- code executes properly
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_plot_ward.py clean of pl
- code executes properly
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_sgd_regression.py clean of pl
- code executes properly
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- bench_tree.py clean of pl
- code executes properly
* docs: removed pylab references from comments
* docs: removed all pylab references
- replaced with matplotlib.pyplot
- pl --> plt
* docs: removed pylab references from comments
- replaced with matplotlib.pyplot
- pl --> plt
* docs: removed all pylab references
- replaced with matplotlib.pyplot
- pl --> plt
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- mlcomp_sparse_document_classification.py clean of pl
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- plot_gpr_noisy_targets.py clean of pl
- code does not execute for extraneous reasons
- File "examples/gaussian_process/plot_gpr_noisy_targets.py", line 31, in
<module>
- from sklearn.gaussian_process import GaussianProcessRegressor
- ImportError: cannot import name GaussianProcessRegressor
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- plot_gpc_isoprobability.py clean of pl
- code does not execute for extraneous reasons
- File "examples/gaussian_process/plot_gpc_isoprobability.py", line 24, in
<module>
- from sklearn.gaussian_process import GaussianProcessClassifier
- ImportError: cannot import name GaussianProcessClassifier
* docs: removed all pylab references
- replaced with matplotlib.pyplot
- pl --> plt
* docs: removed all pylab references
- replaced with matplotlib.pyplot
* refactor: Replace pl with plt
- replace instances of pl (as on import pylab as pl)
with plt (as in import matplotlib.pyplot as plt)
- plot_sparse_coding.py clean of pl
- code executes properly
* docs: removed all pylab references
- replaced with matplotlib.pyplot
* docs: removed all pylab references
- replaced with matplotlib.pyplot
* style: Indent properly
* style: indent properly
* style: Indent properly
* docs: Add missing .pyplot
* docs: Fix typo
* style: Indent properly
* FIX does not draw the ellipse! #6640
* Give ellipses a higher z-order, yellow border, and thicker border.
* Add an alpha level to the markers so that it is easier to tell how
many there are.
* Change false positive to asterisks so that they have similar sizes as
the true positives.
* Fix line too long issue
The comment in a few example plots refer to "the mesh [x_min, m_max] x
[y_min, y_max]" when it should be "x_max".
Also fixed the spacing in plot_random_forest_embedding.py `[x_min,
x_max] x [y_min, y_max]` to `[x_min, x_max]x[y_min, y_max]` to comply
with the coding style of other files with similar code (like those in
the current pull request).
Depreciation of the GMM class.
Modification of the GaussianMixture class.
Some functions from the original GSoC code have been removed, renamed or simplified.
Some new functions have been introduced (as the 'check_parameters' function).
Some parameters names have been changed :
- covars_ -> covariances_ : to be coherent with sklearn/covariances
Addition of the parameter 'warm_start' allowing to fit data by using the previous computation.
The old examples have been modified to replace the deprecated GMM class by the new GaussianMixture class.
Every exemple use the eigenvectors norm to solve the scale ellipse problem (Issues 6548).
Correction of all commentaries from the PR
- Rename MixtureBase -> BaseMixture
- Remove n_features_
- Fix some problems
- Add some tests
Correction of the bic/aic test.
Fix the test_check_means and test_check_covariances.
Remove all references to the deprecated GMM class.
Remove initialized_.
Add and correct docstring.
Correct the order of random_state.
Fix small typo.
Some fix in prevision of the integration of the new BayesianGaussianMixture class.
Modification in preparation of the integration of the BayesianGaussianMixture class.
Add 'best_n_iter' attribute.
Fix some bugs and tests.
Change the parameter order in the documentation.
Change best_n_iter_ name to n_iter_.
Fix of the warm_start problem.
Fix the divergence error message.
Correction of the random state init in the test file.
Fix the testing problems.
Update and add comments into the monotonic test.
Register OneVsRestRegressor as meta estimator
Rename to a more sensible name
Parallel predict and sparse support
Started MultiOutput documentation
Move code to new file multioutput.py
Continuing the move to new multioutput module
Added sample weight support
Better test for sample weights and actually support weights
Added a new test using weighted vs repeated samples to
test sample weight support. Uncovered that weights
were not actually passed on to underlying estimator.
Comment on multiprocess overheads
Move parallel_helper to utils.fixes
This helper works around a python2 limitation on pickling
instance methods
Example of multi-output regression with gradient boosting
Switch to uniform weighted score and updated example
The example now uses a RF with and without the MultiOutput
meta estimator
Added note for removing `score` method
Addressing comments on MultiOutputRegressor
MultiOutputregressor better test for weighted samples
Fix ups
Use explicit keyword argument for passing sample weights and
fix random_state on train-test split in the example
Add gradient calculation in _huber_loss_and_gradient
Add tests to check the correctness of the loss and gradient
Fix for old scipy
Add parameter sigma for robust linear regression
Add gradient formula to robust _huber_loss_and_gradient
Add fit_intercept option and fix tests
Add docs to HuberRegressor and the helper functions
Add example demonstrating ridge_regression vs huber_regression
Add sample_weight implementation
Add scaling invariant huber test
Remove exp and add bounds to fmin_l_bfgs_b
Add sparse data support
Add more tests and refactoring of code
Add narrative docs
review huber regressor
Minor additions to docs and tests
Minor fixes that deals with dealing with NaN values in targets
and old verions of SciPy and NumPy
Add HuberRegressor to robust estimator
Refactored computation of gradient and make docs render properly
Temp
Remove float64 dtype conversion
trivial optimizations and add a note about R
Remove sample_weights special_casing
address @amueller comments
Olivier Grisel