It takes as input a list of tensors of size 2, both of the same shape, and
returns a single tensor, (inputs[[1]] - inputs[[2]]), also of the same
shape.
layer_subtract(inputs, ...)A tensor, the difference of the inputs. If inputs is missing, a
keras layer instance is returned.
A input tensor, or list of two input tensors. Can be missing.
Unnamed args are treated as additional inputs. Named arguments are passed on as standard layer arguments.
Other merge layers:
layer_average(),
layer_concatenate(),
layer_dot(),
layer_maximum(),
layer_minimum(),
layer_multiply()