@fchollet Looks very elegant! One suggestion, in the paper, the content loss is based on comparing activations at conv4_2.
The neural style transfer example for Keras is here: https://github.com/fchollet/keras/blob/master/examples/neural_style_transfer.py …pic.twitter.com/wUMYVM1oYp
-
-
-
@alexjc isn't it also the case here (Except that I use vgg16 instead of 19) - Show replies
New conversation -
-
-
is it possible to force keras to use cpu on tensorflow backend. I use 2GB GTX 960M GPU and get OOM error on TF backend
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.