UPMASK (version 1.2)

innerLoop: UPMASK inner loop

Description

innerLoop executes the UPMASK method's inner loop and returns the stars which were considered as cluster member stars.

The innerLoop perform the PCA, runs the clustering algorithm and check for overdensities in the spatial distribution for the clustered stars in the PC space using a 2d kernel density estimation.

Usage

innerLoop(ocdata_full, ocdata, classAlgol="kmeans", autoThresholdLevel=3, 
autoThreshold=TRUE, iiter=0, plotIter=FALSE, verbosity=1, starsPerClust_kmeans=50, 
nstarts_kmeans=50, runId=0, autoCalibrated=FALSE, stopIfEmpty=FALSE, 
positionDataIndexes=c(1,2), smartTableDB, nDimsToKeep=4, dimRed="PCA", scale=TRUE)

Arguments

ocdata_full

a data frame with the data to perform the analysis

ocdata

a data frame with the data to consider in the PCA step

classAlgol

a string indicating the type of clustering algorithm to consider. Only k-means is implemented at this moment (defaults to kmeans)

autoThresholdLevel

an integer indicating the level for thresholding of the spatial distribution

autoThreshold

a boolean indicating if autoThresolding should be adopted (defaults to TRUE)

iiter

and integer indicating the number of the iteration (passed by the outerLoop)

plotIter

a boolean indicating if the user wants to see iteration plots (defaults to FALSE)

verbosity

a flag indicating the verbosity level: it can be 0 (no screen output at all), 1 (minimum), >=2 (all)

starsPerClust_kmeans

an integer with the average number of stars per k-means cluster

nstarts_kmeans

an integer the amount of random re-initializations of the k-means clustering method (usually it is not necessary to modify this)

runId

an integer greater than zero indicating the run Id (passed by the outerLoop)

autoCalibrated

a boolean indicating if the number of random field realizations for the clustering check in the position space should be autocalibrated (experimental code, defaults to FALSE).

stopIfEmpty

a boolean indicating if the code should completely stop if no spatial clustering is detected (defaults to FALSE)

positionDataIndexes

an array of integers indicating the columns of the data frame containing the spatial position measurements

smartTableDB

a database connection to the smart look-up table

nDimsToKeep

an integer with the number of dimensions to consider (defaults to 4)

dimRed

a string with the dimensionality reduction method to use (defaults to PCA. The only other options are LaplacianEigenmaps or None)

scale

a boolean indicating if the data should be scaled and centered

Value

A data frame with objects considered as members at this iteration.

References

Krone-Martins, A. & Moitinho, A., A&A, v.561, p.A57, 2014

Examples

# NOT RUN {
# Perform a one run of the innerLoop using a simulated open cluster with 
# spatial and photometric data 
# Load the data into a data frame
fileName <- "oc_12_500_1000_1.0_p019_0880_1_25km_120nR_withcolors.dat"
inputFileName <- system.file("extdata", fileName, package="UPMASK")
ocData <- read.table(inputFileName, header=TRUE)
ocData <- data.frame(ocData, id=(1:length(ocData[,1]))) # create an id

# Prepare the data to run the inner loop
posIdx <- c(1,2)
photIdx <- c(3,5,7,9,11,19,21,23,25,27)

# Create the look up table
library(RSQLite)
stcon <- create_smartTable()

# Run the inner loop 
innerLoopRes <- innerLoop(ocData, ocData[,photIdx], autoThresholdLevel=1, verbosity=2,
                          starsPerClust_kmeans=25, positionDataIndexes=posIdx, 
                          smartTableDB=stcon)

# Clean the environment
rm(list=c("inputFileName", "ocData", "posIdx", "photIdx", "innerLoopRes", 
   "fileName"))
dbDisconnect(stcon)
# }
# NOT RUN {
 
# }