On Sat, Nov 12, 2011 at 07:43:08PM +0900, Mathieu Blondel wrote:
> > No, it's not. The data wouldn't take the same memory, as it would be of
> > different size in bytes.
> But it would in theory be possible for int64 / float64 and int32 /
> float32, no?
In terms of size, yes. But if there are oth
Hi all,
I am looking into how to combine classifiers using Scikit-learn.
I think for general purpose, it could be useful to have functions like
stacking and voting in scikit-learn. Is there any plan of developing
ensemble methods?
For now, I am writting my own snippet for stacking. First phase wo
On Sat, Nov 12, 2011 at 6:07 PM, Gael Varoquaux
wrote:
> No, it's not. The data wouldn't take the same memory, as it would be of
> different size in bytes.
But it would in theory be possible for int64 / float64 and int32 / float32, no?
Mathieu
--
On Sat, Nov 12, 2011 at 12:52:07AM +0100, Andreas Mueller wrote:
> afaik it is also possible to convert X from int to float in-place.
No, it's not. The data wouldn't take the same memory, as it would be of
different size in bytes.
Gael
