Error in r_to_py_impl(converted, convert = convert)

[akhst7]

I am having this issue with Leiden running a large an adjacency matrix generated from sNN as follows;

g <- buildSNNGraph(sce, k=20, use.dimred="PCA" ,BPPARAM=MulticoreParam(30), type="jaccard")
g.am<-as_adjacency_matrix(g)
str(g.am)
Formal class 'dgCMatrix' [package "Matrix"] with 6 slots
  ..@ i       : int [1:24346684] 851 1247 1353 2898 3266 3779 4053 4100 4360 4707 ...
  ..@ p       : int [1:99193] 0 189 418 525 607 841 1034 1195 1408 1563 ...
  ..@ Dim     : int [1:2] 99192 99192
  ..@ Dimnames:List of 2
  .. ..$ : NULL
  .. ..$ : NULL
  ..@ x       : num [1:24346684] 1 1 1 1 1 1 1 1 1 1 ...
  ..@ factors : list()

partition<-leiden(g.am)

After about an hour of run, I run into a following error;

Error in r_to_py_impl(converted, convert = convert) : 
  Evaluation error: 351 arguments passed to .Internal(.addCondHands) which requires 5.

if I set sNN k=10, Leiden works just fine but at k=20, it fails.

Any input will be appreciated.