find_peaks: find_peaks

In IncDTW: Incremental Calculation of Dynamic Time Warping

find_peaks

Description

Find negative or positive peaks of a vector in a predefined neighborhood w

Usage

find_peaks(x, w, get_min = TRUE, strict = TRUE)

Arguments

x	vector
w	window, at least w-many values need to be in-between two consecutive peaks to find both, otherwise only the bigger one is returned
get_min	logical (default TRUE) if TRUE, then minima are returned, else maxima
strict	logical, if TRUE (default) then a local minimum needs to be smaller then all neighbors. If FALSE, then a local minimum needs to be smaller or equal all neighbors.

Value

integer vector of indices where x has local extreme values

Examples

#--- Find the peaks (local minima and maxima), 
# and also the border peak at index 29. First the local maxima:
x <- c(1:10, 9:1, 2:11)
peak_indices <- find_peaks(x, w=3, get_min=FALSE)
peak_indices
x[peak_indices]

# and now the local minima
peak_indices <- find_peaks(x, w=3, get_min=TRUE)
peak_indices
x[peak_indices]


#--- What exactly does the neigbohood parameter 'w' mean?
# At least w-many values need to be inbetween two consecutive peaks:
x <- -c(1:10, 9, 9, 11, 9:8, 7)
peak_indices <- find_peaks(x, w=3)
peak_indices
x[peak_indices]

x <- -c(1:10, 9, 9,9, 11, 9:8, 7)
peak_indices <- find_peaks(x, w=3)
peak_indices
x[peak_indices]


#--- What does the parameter 'strict' mean?
# If strict = TRUE, then the peak must be '<' (or '>') 
# then the neighbors, other wise '<=' (or '>=')
x <- c(10:1,  1:10)
peak_indices <-  find_peaks(x, w=3, strict = TRUE)
peak_indices
x[peak_indices]

peak_indices <-  find_peaks(x, w=3, strict = FALSE)
peak_indices
x[peak_indices]
   

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