On 8 November 2015 at 20:42, Sebastian Raschka wrote:
> Hm, I have to think about this more. But another case where I think that the
> handling of categorical features could be useful is in non-binary trees; not
> necessarily while learning but in making predictions more efficiently. E.g.,
> as
Hm, I have to think about this more. But another case where I think that the
handling of categorical features could be useful is in non-binary trees; not
necessarily while learning but in making predictions more efficiently. E.g.,
assuming 3 classes that are perfectly separable by a "color" attr
On 8 November 2015 at 17:50, Sebastian Raschka wrote:
>
>> On Nov 8, 2015, at 11:32 AM, Raphael C wrote:
>>
>> In terms of computational efficiency, one-hot encoding combined with
>> the support for sparse feature vectors seems to work well, at least
>> for me. I assume therefore
>> the problem m
Newton is never d**2 because every body uses a truncated Newton, which is in
effect linear in d.
Gaël
Sent from my phone. Please forgive brevity and mis spelling
On Nov 8, 2015, 18:51, at 18:51, Sebastian Raschka wrote:
>
>> On Nov 8, 2015, at 11:32 AM, Raphael C wrote:
>>
>> In terms of c
> On Nov 8, 2015, at 11:32 AM, Raphael C wrote:
>
> In terms of computational efficiency, one-hot encoding combined with
> the support for sparse feature vectors seems to work well, at least
> for me. I assume therefore
> the problem must be in terms of classification accuracy.
One thing comes
On 5 November 2015 at 13:38, Gael Varoquaux
wrote:
> On Thu, Nov 05, 2015 at 07:05:11AM +, Raphael C wrote:
>> https://github.com/szilard/benchm-ml
>
>> The upshot is that in some cases it seems that the scikit-learn
>> versions have room for improvement.
>
> The various main lessons that I ca
