inference time for this extremely high? about neural-storyteller

[jtoy]

got it all working, needed floatX=float32 ; it went from 70 mins to 6 mins, not bad :)

[YknZhu]

load_all() takes pretty long but it only need to run once. 70 minutes seem to be a bit too long considering you are using GPU. Did you set your .theanorc to run on GPU?

[jtoy]

yeah, I wasnt in gpu mode. I fixed that, but now when I run the same exact code I get:

s = generate.story(z, args.input)

File "/home/ubuntu/experiment/generate.py", line 42, in story
feats = compute_features(z['net'], im).flatten()
File "/home/ubuntu/experiment/generate.py", line 183, in compute_features
deterministic=True).eval())
File "/usr/local/lib/python2.7/dist-packages/lasagne/layers/helper.py", line 191, in get_output
all_outputs[layer] = layer.get_output_for(layer_inputs, *_kwargs)
File "/usr/local/lib/python2.7/dist-packages/lasagne/layers/conv.py", line 257, in get_output_for
conved = self.convolve(input, *_kwargs)
File "/usr/local/lib/python2.7/dist-packages/lasagne/layers/corrmm.py", line 144, in convolve
contiguous_filters = gpu_contiguous(filters)
File "/usr/local/lib/python2.7/dist-packages/theano/gof/op.py", line 611, in call
node = self.make_node(_inputs, *_kwargs)
File "/usr/local/lib/python2.7/dist-packages/theano/sandbox/cuda/basic_ops.py", line 3963, in make_node
input = as_cuda_ndarray_variable(input)
File "/usr/local/lib/python2.7/dist-packages/theano/sandbox/cuda/basic_ops.py", line 47, in as_cuda_ndarray_variable
return gpu_from_host(tensor_x)
File "/usr/local/lib/python2.7/dist-packages/theano/gof/op.py", line 611, in call
node = self.make_node(_inputs, *_kwargs)
File "/usr/local/lib/python2.7/dist-packages/theano/sandbox/cuda/basic_ops.py", line 140, in make_node
dtype=x.dtype)()])
File "/usr/local/lib/python2.7/dist-packages/theano/sandbox/cuda/type.py", line 93, in init
(self.class.name, dtype, name))
TypeError: CudaNdarrayType only supports dtype float32 for now. Tried using dtype float64 for variable None

I dont suppose that is normal? this same code works fine in gpu mode. do i need to convert any of the files or do anything else to make it work in gpu mode?

[jtoy]

Any way I can see the .theanorc file you use? I just force gpu mode with:
THEANO_FLAGS=device=gpu
not sure if I need to tweak other settings.