Hi everyone,
I'm putting together a regression test for a LogisticRegression classifier.
After fixing random_state, I get identical results on linux. But the same
random_state leads to a different result in windows.
Has anyone else noticed this? Any workaround?
Please note that I compare coef_ attribute of the classifier after fitting,
and use numpy.allclose function to compare the floating point numbers. the
discrepancy is not just round-off error. The coefficients are vastly
different. Here's the head of the coef_ attributes:
clf.coef_[0,:5]
Out[16]: array([ 0.52300643, 0.50211767, 1.46571165, -1.67521811,
0.07498549])
clf_true.coef_[0,:5]
Out[17]: array([ 0.52228002, 0.50217637, 1.46595924, -1.6750342 ,
0.07525358])
Best,
Faraz
------------------------------------------------------------------------------
Start Your Social Network Today - Download eXo Platform
Build your Enterprise Intranet with eXo Platform Software
Java Based Open Source Intranet - Social, Extensible, Cloud Ready
Get Started Now And Turn Your Intranet Into A Collaboration Platform
http://p.sf.net/sfu/ExoPlatform
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general