Thank you. It seems that Cholesky uses the GPU only for computing the 
gradient. Am I correct?

On Thursday, November 10, 2016 at 5:00:51 PM UTC+1, nouiz wrote:
>
> I answered on stackoverflow.
>
> On Tue, Nov 8, 2016 at 11:17 AM, Kiuhnm Mnhuik <[email protected] 
> <javascript:>> wrote:
>
>> You can find my question also on stackoverflow: 
>> http://stackoverflow.com/q/40487105/4603642
>>
>>
>> I built a neural network which needs to use the Cholesky decomposition 
>> and solve triangular systems as part of its computation. This means that I 
>> also need to compute the gradients of the whole computation, of course.
>>
>> When I try to compile the code I get the error "No Cula available". 
>> Unfortunately, I can't download Cula from the 
>> http://www.culatools.com/downloads/dense/
>>
>> I was wondering if it's possible to mix CPU and GPU with theano. The 
>> matrices I need to use Cholesky and Solve on are small (100x100) so I could 
>> do that part on the CPU. To do so, though, I'd need to transfer the 
>> matrices to the CPU right after they've been computed by the GPU and then 
>> send the result back to the GPU. Is it possible to do this transparently 
>> enough?
>>
>> -- 
>>
>> --- 
>> 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] <javascript:>.
>> 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.

Reply via email to