topic Re: Multiple y functions as spline parameters in Intel® oneAPI Math Kernel Library

Multiple y functions as spline parameters

kdev1 — Wed, 17 Dec 2025 13:15:52 GMT

Hi,

Is it possible to have multiple y- functions as parameters:

df::spline<double, df::linear_spline::default_type> spl(q);
    spl.set_partitions(partitions.data(), nx)
       .set_coefficients(coeffs.data())
       .set_function_values(functions.data());

eg. can 'functions' be type std::vector<std::vector<double>> or in some other way to compute multiple y's

and if that is not possible how can these splines be computed in parallel?

thank you.

Re: Multiple y functions as spline parameters

Andrey_F_Intel — Thu, 18 Dec 2025 10:49:16 GMT

Hi @kdev1 ,

Thank you for your question.

It's possible. There is the function named set_function_values. It accepts FpType* as input. Follow the documentation link for more information: https://www.intel.com/content/www/us/en/docs/onemkl/developer-reference-dpcpp/2025-2/splines.html .

Let us know if you have more questions

Re: Multiple y functions as spline parameters

Andrey_F_Intel — Thu, 18 Dec 2025 11:03:48 GMT

functions can be std::vector<double> with the size of (nx * ny).

A complete example of using multiple y-functions can be found as part of the oneMKL package.