vector::reserve error during grid_metrics()

[Karltlbm]

Hi JR :)

Creating a grid of classification values, i get this error in the middle of the catalog:

An error occurred when processing the chunk 92. Try to load this chunk with:
 chunk <- readRDS("/tmp/RtmpFAOX6S/chunk92.rds")
 las <- readLAS(chunk)
vector::reserve

Code is:

resolution = 1.0
n_cores = 15
size_tile = 150
size_buffer = 25

path_data = ""
path_data = ""
path_output_local = ""

if(!dir.exists(path_output_local)) dir.create(path_output_local, recursive = TRUE)

ctg <- readLAScatalog(path_data)

opt_chunk_size(ctg)   <- size_tile
opt_chunk_buffer(ctg) <- size_buffer
opt_output_files(ctg) <- file.path(path_output_local, "{ID}_classif")
opt_progress(ctg) <- T

future::plan(future::multisession, workers = n_cores)

majority_class <- function(z) {
  v = which.max(table(z))
  v = as.numeric(names(v))
  return(v)
}

catalog_apply(ctg, function(las_chunk) {
  print(las_chunk)
  r <- grid_metrics(las_chunk, ~majority_class(Classification), res = resolution)
  return(r)}
)

future::plan(future::sequential)
gc()

Do you know what this vector::reserve error is ?

Thanks !

[Jean-Romain]

vector reserve looks like a c++ error. Please share a minimal reproducible example. Maybe you ran out of memory? I see you are using 16 cores. Maybe memory allocation failed.

[Karltlbm]

Here is the problematic chunk

chunk201.zip

chunk <- readRDS("/tmp/RtmpKqRnAy/chunk201.rds")
las <- readLAS(chunk)
Error: vector::reserve

[Jean-Romain]

A LAScluster is just a list that records the bounding box of the current ROI and the path of the corresponding files. Without the files I cannot do anything

class   : LAScluster
name    : noname 
center  : 289750 , 9378550 
extent  : 289700 , 289800 , 9378500 , 9378600 (xmin, xmax, ymin, ymax)
extent+ : 289675 , 289825 , 9378475 , 9378625 (xmin, xmax, ymin, ymax)
size    : 150 x 150 
files   : new_cloud2601813d1b703fe5.laz new_cloud4abb41888e9156cd.laz new_cloud5d5bf1e3078f1219.laz new_cloud5e47389716edf35c.laz new_cloud617cc19b1b9083e4.laz 
filter  : -inside 289675 9378475 289825 9378625