Frédéric Bastien

The Theano benchmark have a sync before the timming too. I don't know if that is useful, but I think consistency between software is better.: ``` theano.sandbox.cuda.synchronize() start = time.time()...

I think it should be available as information somewhere, as for the middle layer it is computed. But maybe we could do 2 summary (fprop + grad weidght) and (fprop...

About the next convolution benchmark, what about 3d convolution for video? On Mon, Sep 8, 2014 at 5:37 AM, Soumith Chintala [email protected] wrote: > nouiz/Theano@092e81e > https://github.com/nouiz/Theano/commit/092e81e009cd531776ef804633f19be1d4f8787a > > —...

Merry Christmas and Happy New Year! (I won't be checking my emails/github frequently until next year, so do not expect fast response) Thanks for the release of fbcufft that will...

@benanne, I agree having the CPU code support all the version is great. But sometimes we need to go faster for research and implementing just the case we use (CUDA)...

Just to note, we also have access to it and have part of it already wrapped in Theano. If you have access to the Nervana code, I can give access...

Just to tell that for Theano, there is a tutorial with a working lstm example there: http://deeplearning.net/tutorial/lstm.html I won't have time to work on it, but I can answer questions...

Same for me. I'll be busy for a few weeks at least. But I can answer question on the code. Le 13 mars 2015 10:59, "Kyunghyun Cho" [email protected] a écrit...

This can be run only between theano fct call. So you won't know the peak. I see 2 ways to do this: 1) give the formula for the extra memory...

I'm pretty buzy, I don't know when I can do it. What do you need exactly, a benchmark with the full forward/backward or just the backward? Is there an example...