@fchollet Looking better! Noticed assert on L86: should work with any resolution of image. The result of tensor dot product is independent.
Playing around with the neural style transfer script for Keras: https://github.com/fchollet/keras/blob/master/examples/neural_style_transfer.py …pic.twitter.com/HvCd2jy4RG
-
-
-
@fchollet As long as that K.dot resolves to Theano's T.tensordot then you can just remove the assert and it should be fine. - Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.