2

I am training a model in Theano 0.9 and Lasagne 0.1 and want to run it on GPU. I've set THEANO_FLAGS as follows:

THEANO_FLAGS=device=gpu0,force_device=True,floatX=float64

Theano prints it is using GPU

Using gpu device 0: GeForce GTX 980 Ti (CNMeM is disabled, cuDNN 4007)

However, I noticed it's not, profiling shows that it's using CorrMM operation which is according to the docs

CorrMM This is a CPU-only 2d correlation implementation taken from caffe’s cpp implementation and also used by Torch.

I have CUDA Toolkit 7.5 installed, Tensorflow works perfectly on GPU. For some reason Theano is falling back to CPU, it is supposed to cause an error due to force_device flag but it's not.

I am not sure where the problem is as I'm new to Theano, I appreciate your help.

talonmies
  • 70,661
  • 34
  • 192
  • 269
yazfield
  • 1,233
  • 11
  • 18

1 Answers1

1

Issue is floatX=float64. Use floatX=float32. GPU supports 32 bit only yet.

Sonam Singh
  • 26
  • 1
  • Thanks! I had to use Float64 because my input data is in that format, The precision shouldn't be a big problem. – yazfield Aug 25 '16 at 02:43