Anders Boesen Lindbo Larsen

Better late than never! :) Your installation is working. The last error is caused by CUDArray lacking broadcasting support for some operations in some rare cases. It should be easy...

Hi! I haven't seen this before. I can't think of any changes that should cause such problems. Have you remembered to clean up the CUDArray folder with both `make clean`...

Thank you for the kind words! I plan on cleaning up the codebase soon and 'releasing' a version 0.1. This might be a good opportunity to put it on PyPI....

Hey, thank you for the info! I have a larger update of the codebase pending and will make sure this is fixed at that point.

This is not a bad idea, will try to do so within a couple of weeks!

@cherryvr: I'm terribly sorry about the lateness of my reply. Due to illness I have been unable to work. Regarding your problem, two ideas spring to mind: 1) I have...

Thank you! :) I have only tried 2D convolutions with DeepPy. I guess faking 1D convolution layers (that is, 2D with one side being 1) should work. Let me know...

Agreed - this is a silly mistake! Thank you for reporting it.

It creates an augmented version of the dataset containin n_augment samples. For the CelebA dataset it doesn't really help and you might simply set n_augment=0 in which disables augmentation.

No, that is not my experience. Are we talking CelebA?