Which version of scikit-learn are you using?
We recently (0.17) removed storing of data point indices in trees which
greatly reduced the size in some cases.
On 04/10/2016 09:28 AM, Piotr Płoński wrote:
Thanks for comments! I put more details of my problem here
http://stackoverflow.com/questio
I am using 0.17.1, did you consider writing custom save methods for this
classifier?
2016-04-11 18:11 GMT+02:00 Andreas Mueller :
> Which version of scikit-learn are you using?
> We recently (0.17) removed storing of data point indices in trees which
> greatly reduced the size in some cases.
>
>
Just curious how it could be made more efficient. ~14.9 Mb for 50 trees on a 20
mb dataset doesn't sound too bad actually since we are not pruning the trees in
Random Forests. Sth I could think would be to summarize similar trees in
buckets or building a "fragment" library of shared decision rul
Yes, there are no doubt more efficient ways to store forests, but it
seems unlikely to be a worthwhile investment.
I think this is a documentation rather than an engineering issue. We
frequently get issues raised that relate to "size": runtime, memory
consumption, model size on disk, (in)effective
Hello Everyone
Me and Andreas were discussing the possibility of a scikit-learn hackathon
during ICML this year. ICML is from June 19 - June 24. How many of you
will be around for it ? And when would you prefer having the hackathon ?
Before or after the conference ?
Thanks
Vighnesh
-
Nice idea! I will be there.
When picking the date, please consider that there are other AI conferences in
NY all around the ICML period (UAI, COLT, IJCAI).
Cheers,
Giorgio
On 12 Apr 2016, at 12:40 PM, Vighnesh Nandan Birodkar
mailto:vighneshbirod...@nyu.edu>> wrote:
Hello Everyone
Me and A
Hi,
Sorry, ICML is at the same dates as the big brain imaging conference, so
I will not be able to attend (neither the conference, nor a sprint).
Cheers,
Gaël
On Mon, Apr 11, 2016 at 10:40:56PM -0400, Vighnesh Nandan Birodkar wrote:
> Hello Everyone
> Me and Andreas were discussing the possibi
