Work around a bug with ndarray.resize(shape) when the shape contain 0.

theano/tensor/elemwise.py

@@ -733,6 +733,11 @@ class Elemwise(Op):
                 # always return an ndarray with dtype object
                 variable = numpy.asarray(variable, dtype=nout.dtype)
             if hasattr(variable,'shape') and storage[0].shape != variable.shape:
+                if numpy.prod(variable.shape) == 0:
+                    # numpy don't resize from a shape (1,5) to (0,5)
+                    # This bypass the inplace... But I it is important in this case.
+                    storage[0] = variable
+                    continue
                 storage[0].resize(variable.shape)
 
             if storage[0].shape: