Does the Machine learning toolbox (decision tree,...) support GPU

Question

Dang Manh Truong 2017-1-25

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/321910-does-the-machine-learning-toolbox-decision-tree-support-gpu

回答： Walter Roberson 2017-1-25

Hi, I'm currently using Matlab R2015a, and I wish to take advantage of the GPU to speedup my code. However, when I used a decision tree and train I received an error: .... = ClassificationTree.fit(data,labels) The variable "data" is of the gpuArray type. So I had to transfer back to CPU to train the classifier. I wonder is there any way I can use gpuArray on training/testing ? And in more recent Matlab releases, are the functionalities GPU-friendly? :(

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Walter Roberson 2017-1-25

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/321910-does-the-machine-learning-toolbox-decision-tree-support-gpu#answer_252000

https://www.mathworks.com/matlabcentral/answers/234776-gpu-computing-for-machine-learning-bagging-ensemble#answer_189972

(In R2015a) "There is no GPU support for decision trees or their ensembles. [...] If you are using 15a or later, trees are multithreaded." (That is, use multiple CPU cores)