Creating sparse matrix from hdf5 file

[rbutleriii]

Hello, I am looking at a large hdf5 file and trying to extract a matrix convert it to a sparse matrix. The source file and code below. It works for smaller slices (i.e [1:5, 1:5]) but when I try to load the whole thing I get an args_regularity_evaluation error

library(hdf5r)
library(Matrix)
fn = "expression_matrix.hdf5"
if(!file.exists(fn)){
  site = "https://idk-etl-prod-download-bucket.s3.amazonaws.com"
  dataset = "aibs_mouse_ctx-hpf_10x"
  # print(paste(site, dataset, i, sep="/"))
  system2("wget", args=paste(site, dataset, fn, sep="/"))
} else { print(paste(fn, "exists, skipping")) }
a = H5File$new(fn, mode="r")
a[['data']]
# b = Matrix(a[['data/counts']][1:5, 1:5], 
           # dimnames=list(a[['data/samples']][1:5], a[['data/gene']][1:5]))
b = Matrix(a[['data/counts']][], 
           dimnames=list(a[['data/samples']][], a[['data/gene']][]))
# Error in args_regularity_evaluation(args = args, ds_dims = simple_extent$dims,  :
#  Number of arguments not equal to number of dimensions: 1 vs. 2

hdf5 file info

Class: H5Group
Filename: /oak/stanford/scg/lab_flongo/reference/allen_brain_atlas/mouse_cortex/expression_matrix.hdf5
Group: /data
Listing:
    name    obj_type    dataset.dims dataset.type_class
  counts H5I_DATASET 1169320 x 31053        H5T_INTEGER
    gene H5I_DATASET           31053         H5T_STRING
 samples H5I_DATASET         1169320         H5T_STRING
   shape H5I_DATASET               2        H5T_INTEGER