Re: [igraph] problems with large graphs

[Tamas Nepusz]

Hi,

> Erdos-Reyni graphs with many vertices [...]
> This seems to happen for various values of m between 0 and n^2 for  
> any n > 249,553
Thanks for reporting that, this is a simple overflow error and it is  
easy to be fixed. In src/graphobject.c in the source of the Python  
interface, locate the following line in  
igraphmodule_Graph_Erdos_Renyi():

if (m < 0 || m > n*n) {

and replace it with this one:

if (m < 0 || ((double)m)/n > n) {

Actually, I think the whole error checking code in this function is  
not necessary, since the parameters are checked once again in the C  
library, but I'll leave it there as is anyway.

> I am running across out of memory errors and segfaults using  
> similarity_inverse_log_weighted() in working with my dataset (~350k  
> nodes, ~4.5m edges).  For my graph and randomly generated graphs of  
> this size, Jaccard and Dice seem to work just fine.  Here is a  
> reproducible example of my problem:
Thanks, I'll dig deeper into that soon.

--
Tamas