Thanks for all the suggestions, I figure out was happening. In R we were
training with all the data(train and test) and later evaluating on test :)
Sorry. Now the results are very similar(0.73 vs 0.74).
Andy, for the results I am using clf.predict_proba.
Thanks.
El 06/10/14 20:13, "Andy" escribi
Hi Zoraida.
I am not expert in R glms but I think the glm call just does logistic
regression.
For the binary case, this is the same as
sklearn.linear_model.LogisticRegression.
Just a wild guess: Did you use clf.decision function results as input to
roc_auc_score?
If you use clf.predict results
thanks Olivier -- much appreciated - this will de-militarize my
conversations in English a lot.
2014-10-06 16:36 GMT+02:00 Olivier Grisel :
> 2014-10-06 15:27 GMT+02:00 Peter Prettenhofer <
> peter.prettenho...@gmail.com>:
> >
> > Both scikit-learn and R (glmnet) should be thoroughly documented.
2014-10-06 15:27 GMT+02:00 Peter Prettenhofer :
>
> Both scikit-learn and R (glmnet) should be thoroughly documented. ML tools
> have come a long way and are very robust and usable these days but they are
> not completely fire-and-forget**.
>
> ** sorry for the military term but I lack a good alter
Hi Zoraida,
can you provide a code snippet (e.g. upload it to gist.github.com) that
illustrates the problem -- especially how you evaluate the goodness of the
predictions (both R and scikit-learn)?
Its pretty difficult to argue about the issue without seeing what you
actually do. The difference be
Hi all,
I know the subject is ugly but I don¹t really know how to call it.
I am newbie with all this machine learning techniques and what I do most
of the time is to follow a ³try and error² approach. I now this method has
some inconvenients but for now
is what I am able to do.
I am working with
