A header-only library for self-tuning functions.
Provides a std::bind like interface for creating self-tuning functions with integer parameters.
// Our function to optimize
auto test_func = [](double x, int ms) noexcept -> double {
std::cout << "Sleeping for " << 10 * ms << " milliseconds" << std::endl;
std::this_thread::sleep_for(std::chrono::milliseconds{10 * ms});
return x * x;
};
// Create our self-tuning function wrapper.
// Every time this function is called it runs a single iteration of the optimization routine
// Forwards the given arguments and current trial values of the paramters to the wrapped function and propagates the return value (if any).
auto func = treble::self_tuning(test_func, std::placeholders::_1,
treble::tunable_param{30/*starting value*/, 0/*min*/, 50/*max*/, 5/*step size*/});
// Counterparts to std::bind_front and std::bind_back are also available
auto func2 = treble::st_back(test_func, treble::tunable_param{30, 0, 50, 5});Details on how to implement your own performance metrics or optimization routines can be found in their respective header files.
When testing performance with a saxpy cuda kernel, no statistically significant overhead was measured once the optimization routine converged.