[scikit-learn] Using GPU in scikit learn

Andreas Mueller t3kcit at gmail.com
Tue Aug 21 09:05:09 EDT 2018


My lab had a GPU library for Random Forests on Images.

It was pretty fast but probably needs some updates:

https://github.com/deeplearningais/curfil


On 8/8/18 10:01 PM, Tommy Tracy wrote:
> Dear Ta Hoang,
>
> Accelerating decision tree ensembles (including Random Forest) is 
> actually a current area of computer architecture research; in fact it 
> is a principle component of my dissertation. Like Sebastian Raschka 
> said, the GPU is not an ideal architecture for decision tree inference 
> because at its core it is a pointer-chasing algorithm (low computation 
> per memory access) that shows low memory locality. Scikit-Learn has 
> done an excellent job with their von Neumann implementation utilizing 
> things like predication and vectorization. If you're looking to go 
> beyond what the CPU can give you, I would point you to FPGAs. If 
> you're interested in discussing this further, let me know.
>
> --
> --
>          Sincerely,
> Tommy James Tracy II
>      Ph.D Candidate
> Computer Engineering
> UniversityofVirginia
>
>
> On Wed, Aug 8, 2018 at 8:50 PM, hoang trung Ta <tahoangtrung at gmail.com 
> <mailto:tahoangtrung at gmail.com>> wrote:
>
>     Dear all members,
>
>     I am using Random forest for classification satellite images. I
>     have a bunch of images, thus the processing is quite slow. I
>     searched on the Internet and they said that GPU can accelerate the
>     process.
>
>     I have GPU NDVIA Geforce GTX 1080 Ti installed in the computer
>
>     Do you know how to use GPU in Scikit learn, I mean the packages to
>     use and sample code that used GPU in random forest classification?
>
>     Thank you very much
>
>     -- 
>     *Ta Hoang Trung (Mr)*
>     *
>     *
>     /Master student/
>     Graduate School of Life and Environmental Sciences
>     University of Tsukuba, Japan
>
>     Mobile:  +81 70 3846 2993
>     Email : ta.hoang-trung.xm at alumni.tsukuba.ac.jp
>     <mailto:ta.hoang-trung.xm at alumni.tsukuba.ac.jp>
>     tahoangtrung at gmail.com <mailto:tahoangtrung at gmail.com>
>     s1626066 at u.tsukuba.ac.jp <mailto:s1626066 at u.tsukuba.ac.jp>
>     *----
>     *
>     /Mapping Technician/
>     Department of Surveying and Mapping Vietnam
>     No 2, Dang Thuy Tram street, Hanoi, Viet Nam
>
>     Mobile: +84 1255151344
>     Email : tahoangtrung at gmail.com <mailto:tahoangtrung at gmail.com>
>
>     _______________________________________________
>     scikit-learn mailing list
>     scikit-learn at python.org <mailto:scikit-learn at python.org>
>     https://mail.python.org/mailman/listinfo/scikit-learn
>     <https://mail.python.org/mailman/listinfo/scikit-learn>
>
>
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scikit-learn/attachments/20180821/09552d33/attachment.html>


More information about the scikit-learn mailing list