The relsa_levels function calculates kmeans-based RELSA levels based on a reference data set.
relsa_levels( refset, mypath, filename = NULL, bsl = bsl, drops = NULL, turns = NULL, relsaNA = NA, k = 4, showScree = "no", customCol = NULL, seed = 123, myYlim = c(0, 1.4), showPlot = "no", saveTiff = "no" )
| refset | normalized data from a reference set |
|---|---|
| mypath | path specification for saving clustering outputs |
| filename | filename for the output *.tiff |
| bsl | baseline object from relsa_baselines function |
| drops | character vector of variables to drop |
| turns | variables with inverted development under severity (i.e. temperature might rise or drop depending on the actual animal model) |
| relsaNA | codes how to handle NaN values during calculations (default is NA) |
| k | the number of k clusters (user defined, default is 4) |
| showScree | show the Scree plot to estimate the optimal number of clusters |
| customCol | custom color string for the k clusters |
| seed | random seed, if set to NULL, no seeding for kmeans is used |
| myYlim | limits of the y-axis for the clustering |
| showPlot | shows the ordered data and colored cluster levels |
| saveTiff | yes/no whether to save the output as *.tiff file (if no, mypath is obsolete) |
levels data.frame with k+1 levels determined by k-means