with newest Theano, libgpuarray, cudnn 6, the following code produces
unexpected results:
x = np.random.randn(100,10,60,60)
k1 = np.random.randn(10,1,5,5)
k2 = np.concatenate([k,k], axis=0)
o1 = T.nnet.conv2d(x,k, num_groups=10)
o1 = o1.eval()
o2 = T.nnet.conv2d(x,k2, num_groups=10)
o2 = o2
Sorry these places where its just k should be k1:
k2 = np.concatenate([k1,k1], axis=0)
o1 = T.nnet.conv2d(x,k1, num_groups=10)
On Thursday, August 10, 2017 at 12:10:45 PM UTC-7, Michael Oliver wrote:
>
> with newest Theano, libgpuarray, cudnn 6, the following code produces
> u
Nevermind: https://github.com/Theano/Theano/issues/6297
On Thursday, August 10, 2017 at 1:56:02 PM UTC-7, Michael Oliver wrote:
>
> Sorry these places where its just k should be k1:
>
> k2 = np.concatenate([k1,k1], axis=0)
>
> o1 = T.nnet.conv2d(x,k1, num_groups=10)
>
>