Re: [igraph] Similarity measures of two vertices out of Memory Igraph

[Tamas Nepusz]

Hi Surendar,

Calculating all the pairwise Jaccard similarities for a graph with  
14798 vertices requires a matrix with 109 475 604 elements. Assuming  
that a double precision floating point value is stored in 8 bytes,  
this matrix would take up roughly 835 MB of memory alone. For  
technical reasons, this matrix will be copied once from the C layer to  
the R layer when the result is ready, so the matrix has to fit in the  
memory of your computer _twice_. You'll need 1.6GB of memory just to  
store the result. I guess that's why you are getting the error message.
What do you intend to do with the similarity values? Maybe we can find  
a workaround if not all the values are needed.

--
Tamas

On 2009.09.22., at 19:29, Surendar Swaminathan wrote:

> Hello All,
>
>  I am trying to find similarity measures for two Vertices and when I
> do similarity.jaccard(g, vids = V(g), mode = c("all"), loops = FALSE)
> I get out of memory error.
>
> Intro of my dataset.What I have undirected weighted Network
>
> sesssionInfo()
>
> R version 2.9.0 (2009-04-17)
> i386-pc-mingw32
>
> locale:
> LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> States.1252;LC_MONETARY=English_United
> States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] igraph_0.5.2-2
>
> Vertices: 14798
> Edges: 122367
> Directed: FALSE
> No graph attributes.
> Vertex attributes: name.
> Edge attributes: weights.
>
> Please help.
>
> This is my graph object
>
> http://dropbox.unl.edu/uploads/20090929/2a22b7fc6b62b21f/Similarity%20Vertices.RData
>
> Is there anywork around for this.
>
>
> Thank you
>
> surendar
>
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help