Created by: juliecbd

when we use paddle.layer.classification_cost(), if the datareader yield python int object, it works fine. But if we enter np.int32 object, this produces error. This may not seem like a big problem, but other cost functions work well with numpy object. So for consistency, I think it would be better to change it to accept np.int32 object

NotImplementedError: Wrong number or type of arguments for overloaded function 'IVector_create'. Possible C/C++ prototypes are: