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
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
update params and minor change
add onlineLDA example
rename onlineLDA to LatentDirichletAllocation
raise NotFittedError in trnasform
change model name in LDA example
cython optimization for online LDA model
fix import error
add evaluate_every parameter in online LDA model
add score in LDA model
fix doc format
update variable name and improve docs
fix broken example and change parameter name to n_samples & n_features
add check_non_negative function in LDA
add parameter to specify online/batch VB in fit method
improve documentation
set default hyper parameter to 1 / n_topics
use gen_batches in example
remove old func from example
n_components < 0 was silently handled like None; > n_features gave a
cryptic error message from the bowels of np.linalg.
XXX n_components == 0 is still allowed as changing it makes tests fail,
not sure what it means.
On current scipy master np.abs(X) triggers:
TypeError: __numpy_ufunc__ not implemented for this type.
I don't know if this is really a bug in scipy or not but in the mean
we can workaround this limitation as this is only for a test anyway.
Lars Buitinck