From f85e82c72e01734f4fbf43a6374a5b2deda24b29 Mon Sep 17 00:00:00 2001
From: nskene <nathan.skene@gmail.com>
Date: Mon, 22 Jul 2019 12:34:16 +0100
Subject: [PATCH] Parallelised optimizeALS

---
 R/liger.R | 76 ++++++++++++++++++++++++++++++++++++++++---------------
 1 file changed, 56 insertions(+), 20 deletions(-)

diff --git a/R/liger.R b/R/liger.R
index 9095b910..c7220eb6 100644
--- a/R/liger.R
+++ b/R/liger.R
@@ -610,10 +610,12 @@ removeMissingObs <- function(object, slot.use = "raw.data", use.cols = T) {
 #' @param V.init Initial values to use for V matrices (default NULL)
 #' @param rand.seed Random seed to allow reproducible results (default 1).
 #' @param print.obj Print objective function values after convergence (default FALSE).
+#' @param num.cores Number of cores to use for optimizing factorizations in parallel (default 1).
 #' @param ... Arguments passed to other methods
 #'
 #' @return \code{liger} object with H, W, and V slots set.
 #' @export
+#' @import parallel
 #'
 #' @examples
 #' \dontrun{
@@ -651,11 +653,13 @@ optimizeALS.list  <- function(
   V.init = NULL,
   rand.seed = 1,
   print.obj = FALSE,
+  num.cores = 1,
   ...
 ) {
   if (!all(sapply(X = object, FUN = is.matrix))) {
     stop("All values in 'object' must be a matrix")
   }
+  
   E <- object
   N <- length(x = E)
   ns <- sapply(X = E, FUN = nrow)
@@ -683,7 +687,43 @@ optimizeALS.list  <- function(
   tmp <- gc()
   best_obj <- Inf
   run_stats <- matrix(data = 0, nrow = nrep, ncol = 2)
-  for (i in 1:nrep) {
+  
+  ### REMOVED THE FUNCTION FROM HERE [for (i in 1:nrep)]
+  if(num.cores==1){
+    outputs = lapply(as.list(1:nrep),FUN=liger_mainSubFunc,rand.seed,g,k,N,ns,W.init,V.init,H.init,lambda,thresh,max.iters,best_obj)
+  }else{
+    require(parallel)
+    cl <- makeCluster(num.cores)
+    outputs = mclapply(as.list(1:nrep),FUN=liger_mainSubFunc,rand.seed,g,k,N,ns,W.init,V.init,H.init,lambda,thresh,max.iters,best_obj,mc.cores=nrep)
+    stopCluster(cl)
+  }
+  
+  for(i in 1:length(outputs)){
+    if (outputs[[i]]$obj < best_obj) {
+      W_m <- outputs[[i]]$W
+      H_m <- outputs[[i]]$H
+      V_m <- outputs[[i]]$V
+      best_obj <- outputs[[i]]$obj
+      best_seed <- outputs[[i]]$seed
+    }
+  }
+  
+  
+  cat("Best results with seed ", best_seed, ".\n", sep = "")
+  out <- list()
+  out$H <- H_m
+  for (i in 1:length(x = object)) {
+    rownames(x = out$H[[i]]) <- rownames(x = object[[i]])
+  }
+  out$V <- V_m
+  names(x = out$V) <- names(x = out$H) <- names(x = object)
+  out$W <- W_m
+  
+  return(out)
+}
+
+liger_mainSubFunc <- function(i,rand.seed,g,k,N,ns,W.init,V.init,H.init,lambda,thresh,max.iters,best_obj){
+  #for (i in 1:nrep)
     set.seed(seed = rand.seed + i - 1)
     start_time <- Sys.time()
     W <- matrix(
@@ -792,13 +832,7 @@ optimizeALS.list  <- function(
       print("Warning: failed to converge within the allowed number of iterations.
             Re-running with a higher max.iters is recommended.")
     }
-    if (obj < best_obj) {
-      W_m <- W
-      H_m <- H
-      V_m <- V
-      best_obj <- obj
-      best_seed <- rand.seed + i - 1
-    }
+    
     end_time <- difftime(time1 = Sys.time(), time2 = start_time, units = "auto")
     run_stats[i, 1] <- as.double(x = end_time)
     run_stats[i, 2] <- iters
@@ -815,17 +849,17 @@ optimizeALS.list  <- function(
     if (print.obj) {
       cat("Objective:", obj, "\n")
     }
-  }
-  cat("Best results with seed ", best_seed, ".\n", sep = "")
-  out <- list()
-  out$H <- H_m
-  for (i in 1:length(x = object)) {
-    rownames(x = out$H[[i]]) <- rownames(x = object[[i]])
-  }
-  out$V <- V_m
-  names(x = out$V) <- names(x = out$H) <- names(x = object)
-  out$W <- W_m
-  return(out)
+    output=list()
+    output$obj = obj
+    output$iters=iters
+    output$end_time=end_time
+    output$run_stats=run_stats[i, ]
+    output$W = W
+    output$H = H
+    output$V = V
+    output$obj = obj
+    output$seed = rand.seed + i - 1
+    return(output)
 }
 
 #' @importFrom methods slot<-
@@ -846,6 +880,7 @@ optimizeALS.liger <- function(
   V.init = NULL,
   rand.seed = 1,
   print.obj = FALSE,
+  num.cores = 1,
   ...
 ) {
   object <- removeMissingObs(
@@ -864,7 +899,8 @@ optimizeALS.liger <- function(
     W.init = W.init,
     V.init = V.init,
     rand.seed = rand.seed,
-    print.obj = print.obj
+    print.obj = print.obj,
+    num.cores = num.cores
   )
   names(x = out$H) <- names(x = out$V) <- names(x = object@raw.data)
   for (i in 1:length(x = object@scale.data)) {