graph-tool September 2016

graph-tool@skewed.de

9 participants
21 discussions

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

How are samples drawn for distance_histogram?
by P-M 30 Sep '16

30 Sep '16

When providing a value for "samples" in "distance_histogram" are the samples drawn randomly? Thus, could I expect a different set of vertices being used if I supply the same graph with the same number of samples twice to compute my distance histogram? I am wondering if I can use the repetition of the computation to give me a crude estimate of the uncertainty due to sampling as opposed to calculating the histogram for all nodes (if the same nodes are being used in each run there is obviously little point in repeating it). I should add that I am using the histogram only to compute the average shortest path length for the network. -- 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 3

inverting boolean PropertyMaps
by Patrick Totzke 26 Sep '16

26 Sep '16

Hi again, I'm using boolean PropertyMaps to represent subsets of graph vertices in combination with `GraphView(g, vfilt=Mask).vertices()` for iteration. I often need the complement of such sets and would like to complement a mask using `numpy.invert`, for efficiency: """ Mask = graph.new_vertex_property('bool') # this fails import numpy notm = numpy.invert(M.ma) NotMask = graph.new_vertex_property('bool', vals=notm) # this works, but is O(|V|) it = imap(lambda x:not Mask[x], graph.vertices()) NotMask = graph.new_vertex_property("bool", vals=it) """ The issue seems to be that in boolean PropertyMap internally uses a PropertyArray of dtype `uint8`, and not `bool`: In the example above (my graph has 6 nodes), Mask.ma is PropertyArray([0, 0, 0, 0, 0, 0], dtype=uint8) So of course, numpy.invert(M.ma) yields PropertyArray([255, 255, 255, 255, 255, 255], dtype=uint8) and not, as expected, the array with only 1's. Is there a reason for using uint8 instead of booleans? And is there a O(1) operation to invert masks? Thanks, Patrick

2 3

GraphView and subclasses of Graph
by Patrick Totzke 21 Sep '16

21 Sep '16

Hi! Thanks for graph-tools, it looks like a fab project! I am toying around with a particular class of graphs for which I wrote a subclass of `graph_tools.Graph`. It should be great to be able to use `GraphView` for filtering (I'm implementing a divide-and-conquer algorithm and want to recur on subgraphs), but it seems that those will always behave as `Graph`, even if I instantiate them with an object of my subclass. Is there a better way to filter my graphs than writing a custom `GraphView` subclass for it? Cheers, Patrick

2 4

Memory error with fit-view and vertex text
by Felix Victor Münch 21 Sep '16

21 Sep '16

Hi, I have a problem with the following code here: https://gist.github.com/FlxVctr/0be83561c6c423222c3fb0799c606667 As soon as I try to fit-view or resize the text I get a memory error even though I have more than 30GB RAM available. Does this functionality rely on GPU memory? Cheers, Felix

3 2

Find minimum value of edge property for all edges connected to a given node
by P-M 16 Sep '16

16 Sep '16

I have a network where each of my edges is labelled with a date. I want to now also label my vertices so that each vertex has a date assigned to it corresponding to the minimum date of all edges incident and emanating from it. Is there an inbuilt function to find this which would be faster than me looping over all vertices and then all edges for each vertex manually? My current code idea is: lowest_year = 2016 for v in g.vertices(): for e in v.in_edges(): year = g.ep.year[e] lowest_year = min(year,lowest_year) for e in v.out_edges(): year = g.ep.year[e] lowest_year = min(year,lowest_year) g.vp.year[v]=lowest_year lowest_year = 2016 -- 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

Need help to locate subpackages
by wish18master 10 Sep '16

10 Sep '16

Hi there, I am new to the graph-tool module. I downloaded the precomplied module from repository and import graph_tool on python. Everything worked out fine, except that I can't use any of the subpackages listed on the documentation (because they were not there). I would really like to use the 'draw' and 'topology' subpackages. Can be someone tell me where to download them or how to import them? Many thanks, Will -- 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 5

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

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

graph-tool September 2016