2014-06-11 18:16 GMT+02:00 Gavin Gray :
> Yeah, you'd have to hand in a ve ctor listing which distribution to use for
> each element in the feature vector. Weka might have a way round this, but
> I'll have to try using it to see what the interface is like. They reference
> a paper that estimates th
Yeah, you'd have to hand in a vector listing which distribution to use for
each element in the feature vector. Weka might have a way round this, but
I'll have to try using it to see what the interface is like. They reference
a paper that estimates the distribution of each feature using KDE:
http://
Hi,
I was wondering if KNN weighs features differently when doing regression
and, if not, would it be possible to do so? I would like to find the
feature weights that minimiz error.
Also, what other kind of pre-processing (such as scaling and normalization)
does KNN do?
Finally, where do I find
Hi,
Did you update Numpy at the same time, by any chance?
There is a discussion on the Numpy ML about a similar message on the
latest beta.
Cheers,
Matthieu
2014-06-11 16:46 GMT+01:00 Miguel Fernando Cabrera :
> Hi Joel, Arnaud,
>
> Thanks for the answer. In fact I am splitting the data using a
>
> Hi Joel, Arnaud,
Thanks for the answer. In fact I am splitting the data using another
approach. Yes I now realize that StratifiedKFold does not make sense here.
But the weird thing is that in 0.14 it does not even complain.
Best Regards,
--
Miguel
--
2014-06-11 15:54 GMT+02:00 Gavin Gray :
> I need to use Naive Bayes for mixed categorial and numerical data and was
> thinking of implementing a flexible Naive Bayes algorithm similar to Weka's
> instead of hacking my way around by converting the numerical to categorical
> or similar. Is there a go
Hi,
I need to use Naive Bayes for mixed categorial and numerical data and was
thinking of implementing a flexible Naive Bayes algorithm similar to Weka's
instead of hacking my way around by converting the numerical to categorical
or similar. Is there a good reason I shouldn't do this? Is anyone el
This has nothing to do with multilabel representation; Stratified K Fold
will not work over multilabel data. But the error message should be clearer.
On 11 June 2014 00:53, Miguel Fernando Cabrera wrote:
> Hi Everyone,
>
> This is my first post in the list. I have been using scikit-learn active
Hi Chris,
your observation is at least partially due to scaling differences between
the losses of the classifiers. Whereas `SGDRegressor` by construction puts
an extra 1/n_samples in front of your data fit term, `Ridge` does not. So
the penalties used will differ by at least a factor n_samples (se
