The lassoLarsCV returns an error if the input matrix is constant
http://jpst.it/y0rf
While this is an extreme situation I think that the algorithm should not
return an error. It should probably throw a warning and then estimate some
sort of model.
In the case of constant input variables the linear regression estimator
returns the constant estimator. I think lasso should do something similar.
At the moment I am planning to do the switch manually. Something like:
if x.std() == 0:
est = constant_model(y)
else:
est = LassoLarsCV().fit(x,y)
is there any class that implement a constant model? Something like f(x) =
mean(y)?
Thanks a lot!
------------------------------------------------------------------------------
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accordance with the BPMN 2 standard
Learn Process modeling best practices with Bonita BPM through live exercises
http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
