Description

visHexMapping is supposed to visualise various mapping items within a supra-hexagonal grid

Usage

visHexMapping(sObj, mappingType = c("indexes", "hits", "dist", "antidist", "bases", 
  "customized"), labels = NULL, height = 7, margin = rep(0.1, 4), area.size = 1, 
      gp = grid::gpar(cex = 0.7, font = 1, col = "black"), border.color = NULL, fill.color = "transparent", 
      lty = 1, lwd = 1, lineend = "round", linejoin = "round", clip = c("on", "inherit", 
          "off"), newpage = TRUE)

Arguments

sObj

an object of class "sMap" or "sInit" or "sTopol"

mappingType

the mapping type, can be "indexes", "hits", "dist", "antidist", "bases", and "customized"

labels

NULL or a vector with the length of nHex

height

a numeric value specifying the height of device

margin

margins as units of length 4 or 1

area.size

an inteter or a vector specifying the area size of each hexagon

gp

an object of class "gpar". It is the output from a call to the function "gpar" (i.e., a list of graphical parameter settings)

border.color

the border color for each hexagon

fill.color

the filled color for each hexagon

lty

the line type for each hexagon. 0 for 'blank', 1 for 'solid', 2 for 'dashed', 3 for 'dotted', 4 for 'dotdash', 5 for 'longdash', 6 for 'twodash'

lwd

the line width for each hexagon

lineend

the line end style for each hexagon. It can be one of 'round', 'butt' and 'square'

linejoin

the line join style for each hexagon. It can be one of 'round', 'mitre' and 'bevel'

clip

either "on" for clipping to the extent of this viewport, "inherit" for inheriting the clipping region from the parent viewport, or "off" to turn clipping off altogether

newpage

logical to indicate whether to open a new page. By default, it sets to true for opening a new page

Value

invisible

Note

The mappingType includes:

• "indexes": the index of hexagons in a supra-hexagonal grid
• "hits": the number of input data vectors hitting the hexagons
• "dist": distance (in high-dimensional input space) to neighbors (defined in 2D output space)
• "antidist": the oppose version of "dist"
• "bases": clusters partitioned from the sMap
• "customized": displaying input "labels"

Examples

# 1) generate data with an iid matrix of 1000 x 9
data <- cbind(matrix(rnorm(1000*3,mean=0,sd=1), nrow=1000, ncol=3),
matrix(rnorm(1000*3,mean=0.5,sd=1), nrow=1000, ncol=3),
matrix(rnorm(1000*3,mean=-0.5,sd=1), nrow=1000, ncol=3))
colnames(data) <- c("S1","S1","S1","S2","S2","S2","S3","S3","S3")

# 2) sMap resulted from using by default setup
sMap <- sPipeline(data=data)

Start at 2018-01-18 16:56:29

First, define topology of a map grid (2018-01-18 16:56:29)...
Second, initialise the codebook matrix (169 X 9) using 'linear' initialisation, given a topology and input data (2018-01-18 16:56:29)...
Third, get training at the rough stage (2018-01-18 16:56:29)...
	1 out of 2 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	2 out of 2 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
Fourth, get training at the finetune stage (2018-01-18 16:56:29)...
	1 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	2 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	3 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	4 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	5 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	6 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
	7 out of 7 (2018-01-18 16:56:29)
	updated (2018-01-18 16:56:29)
Next, identify the best-matching hexagon/rectangle for the input data (2018-01-18 16:56:29)...
Finally, append the response data (hits and mqe) into the sMap object (2018-01-18 16:56:29)...

Below are the summaries of the training results:
   dimension of input data: 1000x9
   xy-dimension of map grid: xdim=15, ydim=15, r=8
   grid lattice: hexa
   grid shape: suprahex
   dimension of grid coord: 169x2
   initialisation method: linear
   dimension of codebook matrix: 169x9
   mean quantization error: 4.27891105682274

Below are the details of trainology:
   training algorithm: batch
   alpha type: invert
   training neighborhood kernel: gaussian
   trainlength (x input data length): 2 at rough stage; 7 at finetune stage
   radius (at rough stage): from 4 to 1
   radius (at finetune stage): from 1 to 1

End at 2018-01-18 16:56:29
Runtime in total is: 0 secs


# 3) visualise supported mapping items within a supra-hexagonal grid
# 3a) for indexes of hexagons
visHexMapping(sMap, mappingType="indexes", fill.color="transparent")
# 3b) for the number of input data vectors hitting the hexagons
visHexMapping(sMap, mappingType="hits", fill.color=NULL)
# 3c) for distance (in high-dimensional input space) to neighbors (defined in 2D output space)
visHexMapping(sMap, mappingType="dist")
# 3d) for clusters/bases partitioned from the sMap
visHexMapping(sMap, mappingType="bases")

Source code

Source man

See also