error while using Jaccard loss

[Dor256]

When i try to train with the given loss the following error accord : ValueError: An operation has None for gradient. Please make sure that all of your ops have a gradient defined (i.e. are differentiable). Common ops without gradient: K.argmax, K.round, K.eval.

[Golbstein]

The Jaccard index isn't a derivable function, therefore you can't use it as a loss.
Try sticking to the categorical_crossentropy or you're welcome to change the jaccard function to be derivable and share the code with us