Reference

Generated

The Generated object represents the front-end class that supports different modes of computing gradients for a provided functor. The modes are: Numerical (finite differencing), CppAD (forward-/reverse-mode AD tracing), and CodeGen (forward-/reverse-mode AD tracing + code generation for various parallel computation platforms).

C++Python

template <template <typename> typename Functor>
struct Generated

In C++, the function to be differentiated/traced needs to be given as a template parameter for a functor type that accepts the Scalar type as template parameter, and implements the operator() function:

void operator()(const std::vector<Scalar>& input,
                std::vector<Scalar>& output) const

Constructor

template <typename... Args>
Generated(const std::string& name, Args&&... args)

The constructor takes the name of the functor and the (optional) arguments to be passed to the constructor of the functor.

import autogen as ag
import numpy as np

def my_function(xs):
    return np.sqrt(xs[0]**2 + xs[1]**2)

def __main__():
    xs = np.array([0.5, 1.5])
    gen = ag.Generated("my_function", my_function)
    gen.set_mode(ag.GENERATE_CUDA)
    y = gen(xs)
    print(y)