fixing nn bug, i hope for good?

R/ComputeEdgelist.R

@@ -73,62 +73,62 @@ ComputeEdgelist <- function(node.object,
   #### MSBR 2024-06-16
   #### Revised MSBR 2025-11-22 to fix k inheritance bug
 
-  if(!(is.null(k)) & nn.method == 'aoz'){
-    # df <- data.frame(x = node.object[[position.x]], y = node.object[[position.y]])
-    # df$barcode <- rownames(df)
-    # df$x <- as.character(df$x)
-    # df$y <- as.character(df$y)
-    # df$x <- as.numeric(df$x)
-    # df$y <- as.numeric(df$y)
-    # df <- df[,c('x','y')] 
-    # 
-    # coords <- as.matrix(df)#cbind(data.list[[i]]$x,data.list[[i]]$y)
-    coords <- as.matrix(cbind(node.object$x,node.object$y))
-    ord <- order(coords[,1])
-
-    n.neighbors <- k # inherits input 'k'
-    n.omp.t <- 1
-
-    n <- nrow(coords)
-
-
-    u.search.type <- 2 ##2 is very fast, 1 is slightly faster than 0, and 0 is the orginal slow one (2 and 0 should match, 1 is also corrected just different opposite sorting among the children)
-
-    indx <- spNNGP:::mkNNIndxCB(coords, n.neighbors, n.omp.t)
-
-    nn.indx <- indx$nnIndx
-    nn.indx.lu <- indx$nnIndxLU
-    nn.indx.run.time <- indx$run.time
-
-    storage.mode(nn.indx) <- "integer"
-    storage.mode(nn.indx.lu) <- "integer"
-
-    n.indx <- spNNGP:::mk.n.indx.list(nn.indx, n, n.neighbors)
-    names(n.indx) <- 1:nrow(coords)
-
-    # build sparse Adjacency Matrix
-    n.indx.names <- names(n.indx)
-    # this unusual code let's me reference a list element's 
-    # name INSIDE the apply function, neat
-    # THIS NEXT STEP AOZ WROTE - IT'S SLOW AND COULD BE SPED UP
-    adj.ij <- lapply(setNames(n.indx.names, n.indx.names), function(nameindex) {
-      data.frame(i=rep(as.integer(nameindex), length(n.indx[[nameindex]])),
-                 j=n.indx[[nameindex]])})
-    adj.ij <- data.table::rbindlist(adj.ij)
-    # clean and make it mutual
-    adj.ij <- na.omit(adj.ij)
-    adj.ij <- rbind(adj.ij, 
-                    data.frame(i=adj.ij$j,
-                               j=adj.ij$i))
-    # map back to original coord indexing
-    adj.ij$i <- ord[adj.ij$i]
-    adj.ij$j <- ord[adj.ij$j]
-    adj.ij <- as.data.frame(adj.ij)
-    colnames(adj.ij) <- c("from", "to")
-    edgelist <- adj.ij
-
-
-  }
+  # if(!(is.null(k)) & nn.method == 'aoz'){
+  #   # df <- data.frame(x = node.object[[position.x]], y = node.object[[position.y]])
+  #   # df$barcode <- rownames(df)
+  #   # df$x <- as.character(df$x)
+  #   # df$y <- as.character(df$y)
+  #   # df$x <- as.numeric(df$x)
+  #   # df$y <- as.numeric(df$y)
+  #   # df <- df[,c('x','y')] 
+  #   # 
+  #   # coords <- as.matrix(df)#cbind(data.list[[i]]$x,data.list[[i]]$y)
+  #   coords <- as.matrix(cbind(node.object$x,node.object$y))
+  #   ord <- order(coords[,1])
+  #   
+  #   n.neighbors <- k # inherits input 'k'
+  #   n.omp.t <- 1
+  #   
+  #   n <- nrow(coords)
+  #   
+  #   
+  #   u.search.type <- 2 ##2 is very fast, 1 is slightly faster than 0, and 0 is the orginal slow one (2 and 0 should match, 1 is also corrected just different opposite sorting among the children)
+  #   
+  #   indx <- spNNGP:::mkNNIndxCB(coords, n.neighbors, n.omp.t)
+  #   
+  #   nn.indx <- indx$nnIndx
+  #   nn.indx.lu <- indx$nnIndxLU
+  #   nn.indx.run.time <- indx$run.time
+  #   
+  #   storage.mode(nn.indx) <- "integer"
+  #   storage.mode(nn.indx.lu) <- "integer"
+  #   
+  #   n.indx <- spNNGP:::mk.n.indx.list(nn.indx, n, n.neighbors)
+  #   names(n.indx) <- 1:nrow(coords)
+  #   
+  #   # build sparse Adjacency Matrix
+  #   n.indx.names <- names(n.indx)
+  #   # this unusual code let's me reference a list element's 
+  #   # name INSIDE the apply function, neat
+  #   # THIS NEXT STEP AOZ WROTE - IT'S SLOW AND COULD BE SPED UP
+  #   adj.ij <- lapply(setNames(n.indx.names, n.indx.names), function(nameindex) {
+  #     data.frame(i=rep(as.integer(nameindex), length(n.indx[[nameindex]])),
+  #                j=n.indx[[nameindex]])})
+  #   adj.ij <- data.table::rbindlist(adj.ij)
+  #   # clean and make it mutual
+  #   adj.ij <- na.omit(adj.ij)
+  #   adj.ij <- rbind(adj.ij, 
+  #                   data.frame(i=adj.ij$j,
+  #                              j=adj.ij$i))
+  #   # map back to original coord indexing
+  #   adj.ij$i <- ord[adj.ij$i]
+  #   adj.ij$j <- ord[adj.ij$j]
+  #   adj.ij <- as.data.frame(adj.ij)
+  #   colnames(adj.ij) <- c("from", "to")
+  #   edgelist <- adj.ij
+  #   
+  #   
+  # }
   return(edgelist)
 
 }

packageUpdateScript.R

@@ -18,6 +18,6 @@ pkgdown::build_article(name = "07 Spatiotemporal NICHES",lazy = T)
 
 # Install from release ?
 install_github('msraredon/NICHES', 
-               ref = 'v1.2.0',
+               ref = 'v1.2.1',
                auth_token = 'f7ea5d8790fe721ac0c9d5ef115d04068b19ed6d',
                force = T)