In the Universal Function object, there are informational attributes that cannot be set:


Attributes
-  __doc__ : A docstring for each ufunc

-  __name__: Name of the ufunc

-  nin: Number of inputs

- nout: Number of output arguments

- nargs: Total number of arguments.

- ntypes: Number of defined inner loops

- types: List of length ntypes containing strings showing types for which the ufunc is defined

- identity: Identity of the ufunc




University of Minnesota - Twin Cities

Essentially, there are NumPy functions that operate on ndarrays. All (ufuncs) wrap a core function that takes scalar inputs and produces scalar outputs.

Why would this be useful?
- Say for example a user wants to sum individual elements in a list with the same corresponding elements in another list. One could iterate through both lists and add them as it goes through the iteration. Or, if the user uses a ufunc called "add(x,y)"  then it does the same thing - but saves times.

Universal Function Object

Universal Function Object Attributes

In the Ufunc object, there are only 4 methods:

- Reduce

- Accumulate

- Reduceat

- outer

However, these methods will only work on ufuncs that take two input arguments and return one output argument. 

Learn Before

Related