What do you need the random matrix for? A baseline?
A simple baseline is to predict the most common class for all obs.
On 15 Mar 2013 14:38, wrote:
> On Fri, Mar 15, 2013 at 10:24 AM, wrote:
> >> Having both margin fixed is an unlikely situation, especially for
> >> confusion matrices. Your ca
Hi
You are correct.
What do you mean by domain of applicability?
The class with the maximum probability gets the predicted discrete value
when you use fit ()
On 19 Mar 2013 14:44, wrote:
> Dear SciKitLearners,
>
> does anyone have experience in using RandomForest's predict_proba function
> as e
How do I make sure the dict is in the same order as my other data?
On 8 March 2013 10:38, Lars Buitinck wrote:
> 2013/3/8 Dirk Nachbar :
> > I want to run some classification and I have some variables which are
> > string, I do not need a bag of words vectorizer, just a simple
Hi
I want to run some classification and I have some variables which are
string, I do not need a bag of words vectorizer, just a simple 10
categories into 10 columns transformation. How do I do that.
BTW thanks to Andreas and Oliver for tutorial last night.
Dirk
--
http://twitter.com/dirknbr
h
