Works fine with CuDNN v5.1 but theano.tensor.grad() crashes python with the current latest cudnn-8.0-windows10-x64-v6.0.
Here is theanoarc: [global] device = cuda0 floatX = float32 optimizer_including=cudnn openmp=False allow_input_downcast=True [cuda] root = C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0 [dnn] enabled = True include_path=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\include library_path=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0\lib\x64 [blas] ldflags=-LC:\\Windows\system32 [gpuarray] preallocate=0.8 -- --- 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.
