This is generally the nature of working in numpy: operations are cheaper
when they're done in bulk.
On 18 November 2014 21:44, Lars Buitinck <[email protected]> wrote:
> 2014-11-18 11:07 GMT+01:00 Nicola Sambin <[email protected]>:
> > - when I computed:
> > for vector in vectors:
> > classifier.predict_proba(vector)
> > it took:
> > 2227,99s user 90,75s system 21:29,94 total
> >
> > - while
> > classifier.predict_proba(vectors)
> > took:
> > 1,06s user 0,39s system 1,984 total
> >
> > What is this (impressive) difference depending on?
>
> Python overhead (slow loops, slow function calls) + memory allocation
> for the slicing (the vectors aren't copied, but a header structure is
> allocated per vector) + memory allocation for the output + input
> validation. If you feed in a batch of vectors, all of them are
> validated in one go, all the loops are performed in C and the output
> matrix is allocated in one operation as well.
>
>
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
>
> http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
> _______________________________________________
> Scikit-learn-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
>
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
_______________________________________________
Scikit-learn-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
