@@ -21,7 +21,9 @@

 #' dat$predicted <- kmeans(dist(dat[,-1]),3)$cluster

 #' # evaluation

 #' evaluateClustering(dat$predicted, dat$cluster)

-evaluateClustering <- function(x, tl){

+evaluateClustering <- function(x, tl=NULL){

+  if(is.null(tl)) tl <- attr(x,"true.labels")

+  if(is.null(tl)) stop("True labels not found!")

   e <- match_evaluate_multiple(x, tl)

   x <- as.character(x)

   unmatched <- length(x)-sum(e$n_cells_matched)

@@ -147,12 +147,14 @@ using the `dr` function.",

       dims <- max(dims,2,na.rm=TRUE)

       x <- get(clustmethod)(x, dims=dims, resolution=resolution, k=k, 

                             steps=steps, min.size=min.size)

-      list( x=x, intermediate_return=evaluateClustering(x,tl) )

+      attr(x, "true.labels") <- tl

+      x

     }

   )

   eva <- list( doublet=NULL, filtering=NULL, normalization=evaluateNorm,

-               selection=NULL, dimreduction=evaluateDimRed , clustering=NULL )

+               selection=NULL, dimreduction=evaluateDimRed, 

+               clustering=evaluateClustering )

   # functions to aggregate the intermediate_return of the different steps

   agg <- list( doublet=.aggregateExcludedCells,

                filtering=.aggregateExcludedCells,