dtwclustControl-class: Class definition for dtwclustControl
Description
Formal S4 class with several control parameters used in dtwclust.
Slots
window.size- Integer or
NULL. Window constraint for DTW, DBA and LB calculations. norm- Character. Pointwise distance for DTW, DBA and the LBs. Either
"L1" for Manhattan distance
or "L2" for Euclidean. Ignored for distance = "DTW" (which always uses "L1") and
distance = "DTW2" (which always uses "L2"). delta- Numeric. Stopping criterion for
DBA centroids and fuzzy clustering. trace- Logical flag. If
TRUE, more output regarding the progress is printed to screen. save.data- Return a "copy" of the data in the returned object? Because of the way
R handles
things internally, all copies should point to the same memory address. symmetric- Logical flag. Is the distance function symmetric? In other words, is
dist(x,y) ==
dist(y,x)? If TRUE, only half the distance matrix needs to be computed. packages- Character vector with the names of any packages required for custom
proxy functions.
See Parallel Computing section in dtwclust. dba.iter- Integer. Maximum number of iterations for
DBA centroids. pam.precompute- Logical flag. Precompute the whole distance matrix once and reuse it at each iteration
if using PAM centroids. Otherwise calculate distances at every iteration.
fuzziness- Numeric. Exponent used for fuzzy clustering. Commonly termed
m in the literature. iter.max- Integer. Maximum number of iterations.
nrep- Integer. How many times to repeat clustering with different starting points. See section
Repetitions in
dtwclust.
Common parameters
-
window.size = NULL
-
norm = "L1"
-
delta = 1e-3
-
trace = FALSE
-
save.data = TRUE
-
symmetric = FALSE
-
packages = character(0)
Only for partitional procedures
-
dba.iter = 15L
-
pam.precompute = TRUE
Only for fuzzy clustering
For both partitional and fuzzy
-
iter.max = 100L
-
nrep = 1L
Details
Default values are shown at the end.