Sampling_via_cluster: Stratified sampling by clusters or types

In bishun945/FCMm: Fuzzy Cluster Method Based on the Optimized m Value (Fuzzifier)

Description

Stratified sampling by clusters or types

Usage

Sampling_via_cluster(
  x,
  num = length(x),
  replace = FALSE,
  order.value = NULL,
  ...
)

Sampling_by_sort(
  x,
  num = length(x),
  log10 = FALSE,
  n_group = 3,
  replace = TRUE
)

Arguments

x	A vector represents the cluster or type, only support numeric value now.
num	The number of sampling. Default as length(x)
replace	The way of sampling. See help(sample) for more details.
order.value	The vector used for ordering. Default is NULL.
...	Parameters of function Sampling_by_sort.
log10	Whether to log10-trans the order.value. Default is FALSE.
n_group	Number of group to be devided. Default is 3.

Value

Result of Sampling_via_cluster is the index of sampled items from the input x.

See Also

Other Algorithm assessment: Assessment_via_cluster(), Getting_Asses_results(), Score_algorithms_interval(), Score_algorithms_sort(), Scoring_system()

Examples

library(FCMm)
x = round(runif(100, 1, 10))
table(x)
w_sampled = Sampling_via_cluster(x, 20)
table(x[w_sampled])


set.seed(1)
N <- 100
x <- stats::rlnorm(N)
ind1 <- Sampling_by_sort(x, N/2)
ind2 <- sample.int(length(x), N/2)

library(ggplot2)
ggplot() + 
geom_density(aes(x = x, color = "Total")) + 
geom_density(aes(x = x[ind1], color = "ind1"), alpha = 0.8) + 
geom_density(aes(x = x[ind2], color = "ind2"), alpha = 0.6) +
scale_x_log10()

bishun945/FCMm documentation built on Oct. 15, 2021, 6:43 p.m.