Thanks a lot for this detailed answer!
Kind regards,
Kevin
Le 14/03/2014 16:37, Olivier Grisel a écrit :
> 2014-03-14 15:34 GMT+01:00 Kevin Keraudren :
>> Hi,
>>
>> I have a question related to the range of my input data for SVM or
>> Random Forests for classification:
>> I normalise my input vec
Manoj, perhaps it's worth making sure this has a regression test.
On 14 March 2014 17:43, Manoj Kumar wrote:
> Yes, I think this has been fixed (intentionally), I remember replacing
> alphas by np.sort(alphas), somewhere in the code.
>
>
> On Fri, Mar 14, 2014 at 5:52 AM, Joel Nothman wrote:
>
Thanks Olivier, I will upload the proposal very soon.
While doing so, I will strengthen my proposal by implementing a basic
version of each of the proposed algorithms, which I will cite in my
proposal.
Cheers. :)
On 3/14/2014 5:38 PM, Olivier Grisel wrote:
> Issam if I am not mistaken you have
hi olivier,
just a question on this statement:
Random Forest (and decision tree-based models in general) are scale
> independent.
>
in many cases with fat data (small samples<50 x many features>10) i
have found that standardizing helps quite a bit in case of extra trees. i
still don't have a
Hi Satra,
In case of Extra-Trees, changing the scale of features might change
the result when the transform you apply distorts the original feature
space. Drawing a threshold uniformly at random in the original
[min;max] interval won't be equivalent to drawing a threshold in
[f(min);f(max)] if f i
thanks gilles,
that makes sense. i haven't checked random forest classification on these
data. i'll check that as well.
cheers,
satra
On Sat, Mar 15, 2014 at 5:51 PM, Gilles Louppe wrote:
> Hi Satra,
>
> In case of Extra-Trees, changing the scale of features might change
> the result when th
2014-03-15 21:53 GMT+01:00 Satrajit Ghosh :
> in many cases with fat data (small samples<50 x many features>10) i have
> found that standardizing helps quite a bit in case of extra trees. i still
> don't have a good understanding as to why this is the case. it could simply
> be small sample bia
Hi arnaud, Eltermann
Thanks for the reply. Firstly I am a noob here so excuse me for any stupid
questions. Eltermann Just my 2 cents, I felt your approach(in BestSplitter
) somewhat inefficient when you are calling the get_sparse_item. Firstly
even though we have indices of all the non zero items,