~/neural-doodle# python3 doodle.py --style /home/mrclon/doodle_test1/style.png  --output /home/mrclon/doodle_test2/new.png --device=cpu --iterations=40
Neural Doodle for semantic style transfer.
  - Using device `cpu` for processing the images.
  - Loading content semantic map from /home/mrclon/doodle_test2/new_sem.png.
  - Loading style image data from /home/mrclon/doodle_test1/style.png.
  - Loading style semantic map from /home/mrclon/doodle_test1/style_sem.png.

Phase #0: resolution 125x83  scale 0.25
  - Style layer sem3_1: 4345 patches in 10,167kb.
  - Style layer sem4_1: 988 patches in 4,588kb.

  0   error 5.04e+03   style 3_1 7.20e+02  4_1 4.17e+03   quality 3.4%
 10   error 2.03e+03   style 3_1 3.40e+02  4_1 1.59e+03   quality 38.3%
 20   error 1.83e+03   style 3_1 2.97e+02  4_1 1.46e+03   quality 59.7%
 30   error 1.75e+03   style 3_1 2.79e+02  4_1 1.42e+03   quality 71.6%

Phase #1: resolution 250x166  scale 0.5
  - Style layer sem3_1: 17920 patches in 41,932kb.
  - Style layer sem4_1: 4345 patches in 20,178kb.

 40   error 2.40e+03   style 3_1 4.75e+02  4_1 1.92e+03   quality 4.5%
 50   error 1.03e+03   style 3_1 1.98e+02  4_1 8.33e+02   quality 41.9%
 60   error 9.51e+02   style 3_1 1.79e+02  4_1 7.71e+02   quality 64.2%
 70   error 9.23e+02   style 3_1 1.71e+02  4_1 7.51e+02   quality 77.3%

Phase #2: resolution 500x333  scale 1.0
Traceback (most recent call last):
  File "/usr/local/lib/python3.4/dist-packages/theano/compile/function_module.py", line 859, in __call__
    outputs = self.fn()
RuntimeError: CorrMM failed to allocate working memory of 576 x 1185600


During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "doodle.py", line 503, in <module>
    generator.run()
  File "doodle.py", line 457, in run
    self.prepare_style(scale)
  File "doodle.py", line 291, in prepare_style
    result = extractor(self.style_image, self.style_map)
  File "/usr/local/lib/python3.4/dist-packages/theano/compile/function_module.py", line 871, in __call__
    storage_map=getattr(self.fn, 'storage_map', None))
  File "/usr/local/lib/python3.4/dist-packages/theano/gof/link.py", line 314, in raise_with_op
    reraise(exc_type, exc_value, exc_trace)
  File "/usr/local/lib/python3.4/dist-packages/six.py", line 685, in reraise
    raise value.with_traceback(tb)
  File "/usr/local/lib/python3.4/dist-packages/theano/compile/function_module.py", line 859, in __call__
    outputs = self.fn()
RuntimeError: CorrMM failed to allocate working memory of 576 x 1185600

Apply node that caused the error: CorrMM{(1, 1), (1, 1)}(Elemwise{Composite{(i0 * (Abs((i1 + i2)) + i1 + i2))}}[(0, 1)].0, Subtensor{::, ::, ::int64, ::int64}.0)
Toposort index: 30
Inputs types: [TensorType(float32, 4D), TensorType(float32, 4D)]
Inputs shapes: [(1, 64, 912, 1300), (64, 64, 3, 3)]
Inputs strides: [(303513600, 4742400, 5200, 4), (2304, 36, -12, -4)]
Inputs values: ['not shown', 'not shown']
Outputs clients: [[Elemwise{Composite{(i0 * (Abs((i1 + i2)) + i1 + i2))}}[(0, 1)](TensorConstant{(1, 1, 1, 1) of 0.5}, CorrMM{(1, 1), (1, 1)}.0, InplaceDimShuffle{x,0,x,x}.0)]]

Backtrace when the node is created(use Theano flag traceback.limit=N to make it longer):
  File "doodle.py", line 503, in <module>
    generator.run()
  File "doodle.py", line 455, in run
    self.model.setup(layers=['sem'+l for l in self.style_layers] + ['conv'+l for l in self.content_layers])
  File "doodle.py", line 163, in setup
    outputs = lasagne.layers.get_output([self.network[l] for l in layers], self.tensor_inputs)
  File "/usr/local/lib/python3.4/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/python3.4/dist-packages/lasagne/layers/conv.py", line 257, in get_output_for
    conved = self.convolve(input, **kwargs)
  File "/usr/local/lib/python3.4/dist-packages/lasagne/layers/conv.py", line 535, in convolve
    filter_flip=self.flip_filters)

HINT: Use the Theano flag 'exception_verbosity=high' for a debugprint and storage map footprint of this apply node.