crp_train: Markov chain Monte Carlo methods for CRP Clustering

In jirotubuyaki/CRPClustering: Bayesian Nonparametric Chinese Restaurant Process Clustering with Entropy

Markov chain Monte Carlo methods for CRP Clustering

Description

Markov chain Monte Carlo methods for CRP Clustering

Usage

crp_train(data = data, alpha = 1, burn_in = 100, iteration = 1000,
  plot = TRUE)

Arguments

data	: a data.frame of data for clustering. Row is each data_i and column is dimensions of each data_i.
alpha	: a numeric of a CRP concentrate rate.
burn_in	: an abandoned iteration integer.
iteration	: an iteration integer.
plot	: a logical type of whether plot a result or not.

Value

result: a list has three elements. The "clusters" is cluster number and joined data number and cluster's mean and variance matrix. The "max" is the cluster number for data i join in. The "z" is the iteration history for each data i join in paticular cluster.

Examples

data <- array(0, dim=c(30, 2))
data[1, ] <- c(-2.43185475495409,2.03311203531621)
data[2, ] <- c(-0.408783769317068,1.75003135626229)
data[3, ] <- c(1.55675133967144,-1.56420523659905)
data[4, ] <- c(0.657368060264562,-0.97383866031164)
data[5, ] <- c(0.068212889245427,1.13418709295999)
data[6, ] <- c(-1.73561815639666,1.81918235905252)
data[7, ] <- c(1.03672457449158,-1.8569658734557)
data[8, ] <- c(1.76100672727452,0.0761038984873157)
data[9, ] <- c(1.65809552931208,-1.68283298969143)
data[10, ] < c(-1.07764013453211,-0.32226716632212)
data[11, ] <- c(-0.261606415434224,1.75394513146391)
data[12, ] <- c(-0.70752394485502,-0.259888201772335)
data[13, ] <- c(-1.38617236009739,-1.24305620615396)
data[14, ] <- c(1.51663782772836,0.0161396983547837)
data[15, ] <- c(-0.914042151747574,1.76495756094862)
data[16, ] <- c(0.282796296711756,-0.0492279088948679)
data[17, ] <- c(1.08831769386705,-0.954851525684512)
data[18, ] <- c(-0.932745904717591,0.762387679372797)
data[19, ] <- c(1.69617665862324,-1.11969182200371)
data[20, ] <- c(-0.767903781929682,1.19342570049071)
data[21, ] <- c(0.401780436172324,-0.457100405625718)
data[22, ] <- c(0.98090840279728,-0.597487493647301)
data[23, ] <- c(-1.29713416781756,0.765401326146141)
data[24, ] <- c(-1.7620625656782,2.02686171889867)
data[25, ] <- c(1.68478051448753,-0.806918914294913)
data[26, ] <- c(0.466622923887724,0.197650126048092)
data[27, ] <- c(1.4851646799543,-1.53289806708663)
data[28, ] <- c(-2.12370907063395,1.6471140089684)
data[29, ] <- c(-0.660332309091402,1.73989289688085)
data[30, ] <- c(-0.957127602683051,-0.0156523076019355)
data <- round(data, 3)
result <- crp_train(
                      data = data,
                      alpha=1,
                      burn_in=10,
                      iteration=100,
                      plot=TRUE
                     )

jirotubuyaki/CRPClustering documentation built on July 5, 2022, 11:33 p.m.