Anders Boesen Lindbo Larsen

Hey, great question and info! I have no idea of how a Celery worker operates. In case a worker spawns a new process and CUDArray somehow doesn't get initialized correctly,...

Hm, according to this [answer on StackOverflow](http://stackoverflow.com/a/33356142), you might want to set your environment variables such that CUDA doesn't get initialized (`export CUDARRAY_BACKEND=NUMPY`). Then, in your worker process, you should...

I just found [this](http://stackoverflow.com/questions/22950047/cuda-initialization-error-after-fork). If Celery creates child processes for workers, then this might be the error you experience. In that case, forcing CUDArray to use NumPy initially, and only...

@jrosebr1: Sorry about the lack of response from my part. I have been unable to work due to illness. From your first message it sounds like an error is preventing...

Ok! Let me know if you run into any error messages.

CUDArray has a NumPy-based backend allowing it to run on CPU. However, the convolution operations on CPU are way too slow to be useful for Neural artistic networks of any...

@mawimawi: You can run without cuDNN if you choose max pooling using the command line option. This might be more entertaining than running on CPU while you are waiting for...

@mawimawi: You are right, better documentation would be nice! :) I'm a bit short on time at the moment, though.

@benwu7: Yes, the NumPy backend could benefit from a more effective convolution implementation. Though, even with fast CPU convolutions, the optimization will probably take around 30 minutes rather than 1-2...

Hey, sorry for the late reply. The convolution operations on CPU are too slow for any practical usage. Sorry about that.