peak()

std::vector< double > MayaFlux::Kinesis::Discrete::peak	(	std::span< const double >	data,
		size_t	n_windows,
		uint32_t	hop_size,
		uint32_t	window_size
	)

Peak amplitude per window.

Parameters

data	Input span
n_windows	Pre-computed window count from num_windows()
hop_size	Samples between window starts
window_size	Samples per window

Returns

Per-window peak absolute values

Definition at line 51 of file Analysis.cpp.

{
    std::vector<double> out(n_windows);
    std::vector<size_t> idx(n_windows);
    std::iota(idx.begin(), idx.end(), 0);
 
    Parallel::for_each(Parallel::par_unseq, idx.begin(), idx.end(),
        [&](size_t i) {
            const size_t start = i * hop_size;
            auto w = data.subspan(start, std::min<size_t>(window_size, data.size() - start));
            double mx = 0.0;
            for (double s : w)
                mx = std::max(mx, std::abs(s));
            out[i] = mx;
        });
 
    return out;
}

Referenced by cross_correlate().

Here is the caller graph for this function: