Lars,
I am using Gaussian Bayes, Bernoulli Bayes, k-Nearest Neighbors, and the
Decision Tree. None of them are taking string classes, hence why I was
using the LabelEncoder. Maybe I have an old/broken version? I have
scikit-learn version 0.14.
On Wed, Jul 17, 2013 at 5:30 PM, <
[email protected]> wrote:
>
> 2013/7/17 Wifi Gi <[email protected]>:
> > I'm aware of how to use LabelEncoder to convert string classes to numeric
> > and back again. However, in my applications, I frequently save my models
> to
> > disk (by pickling them) and reload them when I want to use them. When I
> use
> > predict(), it of course returns only numerical classes, and if you don't
> > remember what the numbers stood for, it's useless. Is there a way to
> somehow
> > include a LabelEncoder in with the classifier object model when it is
> saved,
> > so I don't have to write the class encodings to a separate file somewhere
> > else?
>
> Most classifiers will actually take string class labels, so you may
> not need a LabelEncoder. Which classifier are you using?
>
> (All classifiers should take string labels, but I'm not sure if were
> implemented it for all of them. This feature was added as a bit of an
> afterthought.)
>
> --
> Lars Buitinck
> Scientific programmer, ILPS
> University of Amsterdam
>
>
> End of Scikit-learn-general Digest, Vol 42, Issue 59
> ****************************************************
>
------------------------------------------------------------------------------
See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
