August 2016 - graph-tool - archives.skewed.de

Local Clustering Coefficient for k<=1
by P-M 23 Oct '16

23 Oct '16

How does graph-tool deal with vertices having a degree of 1 (or 0) when calculating the local clustering coefficient graph_tool.clustering.local_clustering? Given the definition used this value would be undefined for k<=1. -- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com… Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.

3 17

Re: [graph-tool] Meaning of vertex_corr in block stochastic?
by Tiago Peixoto 03 Oct '16

03 Oct '16

Sorry for the late reply, but there was a problem with the mailing list. ttfoley wrote > I can't figure out the precise meaning of 'proportional' to, and would > like to know where/how it enters into the the actual formulae for > stochastic blocks. I have tried playing with various limits, and of course > tried scaling the 'probability' to account for graph size. For example, > there does not appear to be a limit where all edges are within blocks, > though surely this should be possible(assuming the number of edges fits > within the blocks)? Any help would be greatly appreciated. Proportional means exactly that: The probability will be what you give, up to an overall constant, which is unimportant since the total number of edges is always fixed. To force the edges to always be inside the groups, you should set the probability to zero, whenever r != s. ----- -- Tiago de Paula Peixoto <tiago(a)skewed.de> -- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com… Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.

3 2

kcore_decomposition using out degree
by Alexandre Bovet 05 Sep '16

05 Sep '16

Dear all, I am wondering how exactly the k-core decomposition for directed network is defined in graph-tool. With the following network: Gt = gt.Graph(directed=True) Gt.add_edge_list([(1, 0),(2, 0),(2, 1),(3, 1),(3, 0),(4, 0),(4, 1),(5, 1), (5, 0),(6, 0),(6, 1),(7, 4),(7, 0),(8, 3),(8, 0),(9, 0), (9, 2),(10, 2),(10, 3),(11, 2),(11, 3),(12, 5),(12, 3), (13, 2),(13, 12),(14, 11),(14, 0)]) The k-core decomposition is: print(gt.kcore_decomposition(Gt, 'in').a) >>> [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0] print(gt.kcore_decomposition(Gt, 'out').a) >>> [10 5 4 4 1 1 0 0 0 0 0 1 1 0 0] print(gt.kcore_decomposition(Gt, 'total').a) >>> [2 2 2 2 2 2 2 2 2 2 2 2 2 2 2] I don't understand the results for the out-k-core. The same analysis in igraph (using shell_index()) gives: in-core : [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] out-core : [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0] total : [2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2] Looking at the source code of kcore_decomposition, it looks like when deg=='out', the decomposition is done on the reversed network, but still using the out-degree. Is it normal? I am a bit confused. Thanks, Alex

2 2

Average path length - shortest_distance=2147483647
by P-M 05 Sep '16

05 Sep '16

I am trying to find the average shortest path length for my network. I am currently trying to find it by running dist = gt.shortest_distance(g) and then finding the average of the results. I however find that for a lot of distances I get the value "2147483647" which is seemingly imposisbly large for my network and crops up in the example results in the documentation too (I only have 718 vertices and 979 edges in my graph). Does this value have a special significance? Does it mean there is no path between the given vertices? Also, seeing as there is no reference to it in the documentation I presume that this algorithm has not been parallelised? -- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com… Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.

3 2

Shortest_distance complexity when used with max_dist
by François 05 Sep '16

05 Sep '16

<disclaimer> I hope this question isn't too dumb, if it is please accept my apologies. </disclaimer> Hi list, The graph that I work with has ≈ 30M vertices and ≈ 60M edges. This graph has positive weights, it represents a Road network. I need to compute the shortest path form a node to any other node within a range. The documentation shortest_distance <http://graph-tool.skewed.de/static/doc/topology.html#graph_tool.topology.sh…> , states that the complexity (source and weights specified) is O(VlogV). My question is: "What happens when the *max_dist* parameter is specified ?" I observed that the execution time of *shortest_distance* increases linearly with the number of vertices. The execution time goes from 19ms with a sub-sample of 1M vertices and 2M edges to 320ms with the full graph. How can it be explained ? Thanks for your help, François. -- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com… Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.

2 17

Graph drawing using edges weights and cluster labels.
by Александр Салынский 31 Aug '16

31 Aug '16

Hello, I've tryed to use graph-tool for solving my problem of visualizing some set of data using correlarions matrix filled with float values from the interval [0, 1] where 0 matches to non-correlated at all values and 1 for the same values. So now I've faced with problem, that I coudn't draw the graph with respect to this values(each time I've received smth like sphere with overlayed values), also I have some pre-computed clusterisations so I also would like to try to visualize the graph using them as the labels of the class for each vertex. So I'll be very pleased for anyone help me with some information about how to: 1) Visualize the graph with respect to the adjacency matrix(filled with values represents correlations of each vertex pairs). 2) Visualize the graph with respect to the labels of the cluster for each vertex. Best regards, Salynskii Aleksandr. P.S. I understand, that answers for my questions mb somewhere in the documentation, but coundn't find ways to solve this problems, so hope that someone have any links of samples(even better) to help me with it.

1 0

conda package anyone?
by Felix Victor Münch 21 Aug '16

21 Aug '16

Hi all, is anybody aware, or planning to create, a conda package of the latest version of graph-tool? I tried to compile it myself but ran into a lot of issues that I could not solve due to my mediocre knowledge in doing so. Cheers, Felix

3 3

positions of vertices
by Глеб Киселев 17 Aug '16

17 Aug '16

Good day to you. I'm trying to create pop-up windows in graph-tool graph. Using your advises I add a function, which making callback from the graph and give coordinates of my mouse pointer. But graph-vertices coordinates are not simular with gdk-window coordinates. In example: I placed some vertices, but a vertex with (0,0) coordinates not in left-up angle of the window. How I can get coordinates of vertices in gdk style? My pointer's coordinates of (0,0) are in left-up side of the window.

1 0

Average Clustering Coefficient Distribution
by P-M 15 Aug '16

15 Aug '16

Hi, I am interested in computing the distribution of clustering coefficient with degree, c(k), for a network of mine (c.f. e.g. p. 102 in dx.doi.org/10.1038/nrg1272 ). Before I go and write a script to calculate this using the local_clustering method implemented in graph-tool I was wondering if anybody knew of any methods implemented to this end already? -- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com… Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.

2 2

deepcopy of Vertices, Edges and Graphs
by Evangelos Petsalis 15 Aug '16

15 Aug '16

Hi, After processing an external text file I create a an object that among other things holds a graph object, properties, vertices etc etc. I would like to save the original copy of the object, so after I can always refer back to it in a speedy fashion without having to re-parse the text file and re-initialize the object, but when I tried to use deepcopy() it complains. I can not even perform a deep copy on a Vertex object. Any idea why this is happening, and most importantly is there a work around? (I tried pickle and shelve and they do not work either). Thanks in advance for the response, Vaggelis -- View this message in context: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com… Sent from the Main discussion list for the graph-tool project mailing list archive at Nabble.com.

2 1