dtw2vec: Fast vector-based Dynamic Time Warping

In IncDTW: Incremental Calculation of Dynamic Time Warping

Fast vector-based Dynamic Time Warping

Description

Calculates the Dynamic Time Warping distance by hand of a vector-based implementation and is much faster than the traditional method dtw(). Also allows early abandoning and sakoe chiba warping window, both for univariate and multivariate time series.

Usage

dtw2vec(Q, C, dist_method = c("norm1", "norm2", "norm2_square"),
         step_pattern = c("symmetric2", "symmetric1"),
         ws = NULL, threshold = NULL)

Arguments

Q	Either Q is (a) a time series (vector or matrix for multivariate time series) or (b) Q is a cost matrix, so a matrix storing the local distances of the time series Q and C. If Q and C are matrices, they need to have the same number of columns. If Q is a cost matrix, C needs to be equal the character string "cm".
C	time series as vector or matrix, or for case (b) C equals "cm"
dist_method	character, describes the method of distance measure. See also dtw. If Q is a cost matrix, the dist_method parameter is not necessary.
step_pattern	character, describes the step pattern. See also dtw.
ws	integer, describes the window size for the sakoe chiba window. If NULL, then no window is applied. (default = NULL)
threshold	numeric, the threshold for early abandoning. In the calculation of the global cost matrix a possible path stops as soon as the threshold is reached. Facilitates faster calculations in case of low threshold. The threshold relates to the non-normalized distance measure. (default = NULL, no early abandoning)

Details

no matrices are allocated, no matrices are returned

Value

distance	the DTW distance
normalized_distance	the normalized DTW distance, see also dtw

References

• Leodolter, M.; Pland, C.; Brändle, N; IncDTW: An R Package for Incremental Calculation of Dynamic Time Warping. Journal of Statistical Software, 99(9), 1-23. doi: 10.18637/jss.v099.i09

• Sakoe, H.; Chiba, S., Dynamic programming algorithm optimization for spoken word recognition, Acoustics, Speech, and Signal Processing [see also IEEE Transactions on Signal Processing], IEEE Transactions on , vol.26, no.1, pp. 43-49, Feb 1978. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1163055

Examples

Q <- cumsum(rnorm(100))
C <- Q[11:100] + rnorm(90, 0, 0.5)
dtw2vec(Q = Q, C = C)
dtw2vec(Q = Q, C = C, ws = 30)
dtw2vec(Q = Q, C = C, threshold = 100)
dtw2vec(Q = Q, C = C, ws = 30, threshold = 100)

cm0 <- cm(Q, C)
dtw2vec(Q = cm0, C = "cm", ws = 30, threshold = 100)

IncDTW documentation built on March 18, 2022, 6:43 p.m.