* insert versionadded versionchanged directives in docstrings for 0.18
indicate where exception classes were moved from
* moved versionadded in the proper places
Ensuring consistent transforms for KernelPCA
Taking @vene's changes into account, thanks!
Taking @jakevdp's comment into account
Added more verbose documentation to kernel_pca.py
Specifying that X_fit_ will not be None. @jakevdp
KernelPCA: Fixing more formatting of docstring
Addressed @vene's documentation comments
Addressing that dual_coef_ might not be present in model in docs. @vene
Update kneighbors_graph()
Add n_job attribute.
Update Isomap
Add n_job attribute to Isomap.__init__ method.
Conform isomap.py to pep8.
Conform graph.py to pep8.
Update dbscan_.py
Add n_jobs parameter to dbscan function and DBSCAN.__init__ method.
Refactor DBSCAN.__init__ documentation.
Update mean_shift_.py
Add n_jobs parameter to estimate_bandwidth function.
Propagate existing n_jobs parameter in mean_shift to estimate_bandwidth and NearestNeighbors.
Conform mean_shift_.py to pep8.
Conform locally_linear.py to pep8.
Add n_jobs parameter to SpectralClustering.__init__ method.
Add n_jobs parameter to SpectralEmbedding.__init__ method.
Update usage of barycenter_kneighbors_graph function.
Conform spectral_embedding_.py to pep8.
Update label_propagation.py
Conform to pep8.
Refactor imports.
Add n_jobs parameter to LabelSpreading.__init__ method.
Update dbscan function call.
Remove n_jobs argument from dbscan call, as it is already being given by self.get_params().
Add n_jobs parameter to KernelPCA.__init__ method.
Update KernelPCA construction in Isomap.
Remove unnecessary parentesis in isomap.py:211,29.
Update docs: pipeline.rst
Revert "Fix n_jobs usage"
This reverts commit 9480630a69bbb68b01d1176d476222fc665cdd1e.
Fix usage of n_jobs property in Isomap
`v0 = random_state.rand(M.shape[0])` leads to an initial residual vector in ARPACK which is all positive. However, this is not the absolute or squared residual, but a true difference. Thus, it is better to initialize with `v0=random_state.uniform(-1, 1, M.shape[0])` to have an equally distributed sign. This is the way that ARPACK initializes the residuals.
The effect of the previous initialization is that eigsh frequently does not converge to the correct eigenvalues, e.g. negative eigenvalues for s.p.d. matrix, which leads to an incorrect null-space.
- initialized all occurences of sklearn.utils.arpack.eigsh the same way it would be initialzed by ARPACK
- regression test to test behavior of new initialization
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
Replaced all BSD style licenses with "BSD 3 clause"
Not checked yet!
Removed duplicate "3 clause, 3 clause"
Removed trailing period if exists
Fixed some missed licences, still about 50 to do, but those can be automated
Think I got the last of them.
Apparently me and sed have different ideas of regex.
Found a few more
Adrin Jalali
Hanmin Qin