Use more recent Theano like the newly released 0.9 version. Concatenate should not take 2s. If you still have slow speed after the update, give codes that reproduce that and show how you do your timing. You should not include the compilation time in your timing.
On Wed, Mar 22, 2017 at 7:32 AM <[email protected]> wrote: > Hi guys: > > I want to concatenate two tensor so I'm using the concatenate to do > that. > However, I found it takes 2s to finish one concatenate operation no > matter the size of tensor. > It's especially time unacceptable especially when incorporating > multiple concatenation in my code. > Is there any way to solve or replace this operation? > > By the way, my gpu is TITAN X with cnmem = 0.75, the cuDNN version is > 4004, the cuda version is 7.0 and theano version is 0.82 > > Thanks > Xiao > > -- > > --- > You received this message because you are subscribed to the Google Groups > "theano-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/d/optout. > -- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
