A collection of static helper functions for dealing with arrays.
This is a wrapper class that can hold two double arrays, one of which is interpreted as containing the real values, the other the imaginary values of the complex numbers.
A complex number: real+sqrt(-1).imag
Dynamic programming to compute similarity measure
FFT for non-power-of-two sequences Note that, this algorithm is significantly slower than FFT.java.
FloatList is used to maintain a circular buffer of float values.
Computes the N-point Discrete Hilbert Transform of real valued vector x: The algorithm consists of the following stages: - X(w) = FFT(x) is computed - H(w), DFT of a Hilbert transform filter h[n], is created: H=H[N/2]=1 H[w]=2 for w=1,2,...,N/2-1 H[w]=0 for w=N/2+1,...,N-1 - x[n] and h[n] are convolved (i.e.
Principal component analysis solve PCA using eigenvectors decomposion and singular value decomposition (SVD).
Multiple linear regression For the case of k independent variables x_1, x_2, ...
Copyright © 2000–2016 DFKI GmbH. All rights reserved.