warnings.warn("The default axis of MaxAndArgmax will change! Now we return the max and the armax over the last dimensions. It will change to be the same as numpy: the max and argmax over all dimensions. To hide this warning and be compatible with the future behavior, set axis to -1 to have the current behavior. MaxAndArgmax currently support axis over only 1 dimensions, so you must flatten the tensor to have the futur behavior.")
...
...
@@ -1676,7 +1679,7 @@ def max(x, axis='DEFAULT'):
:note: we return an error as numpy when we reduce a dim with a shape of 0
:note2: see MaxAndArgmax note for a difference between numpy and theano when axis==None