This package uses forward automatic differentiation to calculate the first and second derivatives of provided user functions. It is 25-30 times faster than
commercially available packages (June 2006).
To calculate the derivatives of function myFunc with respect to x at x=x0 you call
x = myAD(x0);
result = myFunc(x);
functionValue = getvalues(result);
derivatives = getderivs(result);
I am getting this error, do you have any idea of what can be happening?
The following error occurred converting from myAD to double:
Error using double
Conversion to double from myAD is not possible.
I run into the same problem as described by David, input_variable(1)*internal_function_vector(1:N) couldn't go through, the operation is not covered. Otherwise it'll be a terrific package.
nice and usefull
This package is only useful if there are no matrix operations in your function, and this includes repmat on vectors. An operation as simple as
est = input_variable(1) * internal_function_vector(1:N)
falls outside the bounds of this routine, so be warned!
how to use this?plz, help me
This is great! Thanks so much. Would you like to port it to R?
is very insteresting