Ask about CPU version of ''Generative Adversarial Networks" Code

[yzho...@berkeley.edu]

Hi I'm a student from Berkeley and I just read the paper ''Generative Adversarial Networks". I wanted to try out some of your code for a project but I cannot use the version from the github for the lack of GPU support. As suggested by the running error, is it ok for me to get a CPU version of it? It would help a lot. Thanks very much.

[Frédéric Bastien]

You can probably modify that code to run on CPU. Check the error and modify the corresponding part. Now we probably have the CPU op corresponding to the GPU op used in that paper.

Fred

On Fri, Apr 22, 2016 at 12:34 AM, <yzho...@berkeley.edu> wrote:

Hi I'm a student from Berkeley and I just read the paper ''Generative Adversarial Networks". I wanted to try out some of your code for a project but I cannot use the version from the github for the lack of GPU support. As suggested by the running error, is it ok for me to get a CPU version of it? It would help a lot. Thanks very much.

--

---
You received this message because you are subscribed to the Google Groups "pylearn-dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pylearn-dev...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[yzho...@berkeley.edu]

Ok thanks!

[wxwan...@gmail.com]

Nice to see you, I'm also a student who majored in CS and I download the source code of GAN from "github.com/goodfeli/adversarial", I wonder if you have solved the problem, following the guide, I have modify the code of pylearn2/models/maxout, which has a code of CPU version.

However, another problem occurs, adversarial/deconv.py also need a GPU and report the same error, which I don't know what to do.

If you have run GAN code well, could you tell me how could you solve the problem? Thank you !

在 2016年4月24日星期日 UTC+8上午9:02:12，yzho...@berkeley.edu写道：