graph_tool.stats
 Miscellaneous statistics¶
Summary¶
vertex_hist 
Return the vertex histogram of the given degree type or property. 
edge_hist 
Return the edge histogram of the given property. 
vertex_average 
Return the average of the given degree or vertex property. 
edge_average 
Return the average of the given degree or vertex property. 
label_parallel_edges 
Label edges which are parallel, i.e, have the same source and target vertices. 
remove_parallel_edges 
Remove all parallel edges from the graph. 
label_self_loops 
Label edges which are selfloops, i.e, the source and target vertices are the same. 
remove_self_loops 
Remove all selfloops edges from the graph. 
remove_labeled_edges 
Remove every edge e such that label[e] != 0. 
distance_histogram 
Return the shortestdistance histogram for each vertex pair in the graph. 
Contents¶

graph_tool.stats.
vertex_hist
(g, deg, bins=[0, 1], float_count=True)[source]¶ Return the vertex histogram of the given degree type or property.
Parameters: g :
Graph
Graph to be used.
deg : string or
PropertyMap
Degree or property to be used for the histogram. It can be either “in”, “out” or “total”, for in, out, or total degree of the vertices. It can also be a vertex property map.
bins : list of bins (optional, default: [0, 1])
List of bins to be used for the histogram. The values given represent the edges of the bins (i.e. lower and upper bounds). If the list contains two values, this will be used to automatically create an appropriate bin range, with a constant width given by the second value, and starting from the first value.
float_count : bool (optional, default: True)
If True, the counts in each histogram bin will be returned as floats. If False, they will be returned as integers.
Returns: counts :
ndarray
The bin counts.
bins :
ndarray
The bin edges.
See also
edge_hist
 Edge histograms.
vertex_average
 Average of vertex properties, degrees.
edge_average
 Average of edge properties.
distance_histogram
 Shortestdistance histogram.
Notes
The algorithm runs in \(O(V)\) time.
If enabled during compilation, this algorithm runs in parallel.
Examples
>>> from numpy.random import poisson >>> g = gt.random_graph(1000, lambda: (poisson(5), poisson(5))) >>> print(gt.vertex_hist(g, "out")) [array([ 7., 33., 91., 145., 165., 164., 152., 115., 62., 29., 28., 6., 1., 1., 0., 1.]), array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16], dtype=uint64)]

graph_tool.stats.
edge_hist
(g, eprop, bins=[0, 1], float_count=True)[source]¶ Return the edge histogram of the given property.
Parameters: g :
Graph
Graph to be used.
eprop :
PropertyMap
Edge property to be used for the histogram.
bins : list of bins (optional, default: [0, 1])
List of bins to be used for the histogram. The values given represent the edges of the bins (i.e. lower and upper bounds). If the list contains two values, this will be used to automatically create an appropriate bin range, with a constant width given by the second value, and starting from the first value.
float_count : bool (optional, default: True)
If True, the counts in each histogram bin will be returned as floats. If False, they will be returned as integers.
Returns: counts :
ndarray
The bin counts.
bins :
ndarray
The bin edges.
See also
vertex_hist
 Vertex histograms.
vertex_average
 Average of vertex properties, degrees.
edge_average
 Average of edge properties.
distance_histogram
 Shortestdistance histogram.
Notes
The algorithm runs in \(O(E)\) time.
If enabled during compilation, this algorithm runs in parallel.
Examples
>>> from numpy import arange >>> from numpy.random import random >>> g = gt.random_graph(1000, lambda: (5, 5)) >>> eprop = g.new_edge_property("double") >>> eprop.get_array()[:] = random(g.num_edges()) >>> print(gt.edge_hist(g, eprop, linspace(0, 1, 11))) [array([501., 441., 478., 480., 506., 494., 507., 535., 499., 559.]), array([0. , 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ])]

graph_tool.stats.
vertex_average
(g, deg)[source]¶ Return the average of the given degree or vertex property.
Parameters: g :
Graph
Graph to be used.
deg : string or
PropertyMap
Degree or property to be used for the histogram. It can be either “in”, “out” or “total”, for in, out, or total degree of the vertices. It can also be a vertex property map.
Returns: average : float
The average of the given degree or property.
std : float
The standard deviation of the average.
See also
vertex_hist
 Vertex histograms.
edge_hist
 Edge histograms.
edge_average
 Average of edge properties.
distance_histogram
 Shortestdistance histogram.
Notes
The algorithm runs in \(O(V)\) time.
If enabled during compilation, this algorithm runs in parallel.
Examples
>>> from numpy.random import poisson >>> g = gt.random_graph(1000, lambda: (poisson(5), poisson(5))) >>> print(gt.vertex_average(g, "in")) (4.96, 0.0679882342762334)

graph_tool.stats.
edge_average
(g, eprop)[source]¶ Return the average of the given degree or vertex property.
Parameters: g :
Graph
Graph to be used.
eprop :
PropertyMap
Edge property to be used for the histogram.
Returns: average : float
The average of the given property.
std : float
The standard deviation of the average.
See also
vertex_hist
 Vertex histograms.
edge_hist
 Edge histograms.
vertex_average
 Average of vertex degree, properties.
distance_histogram
 Shortestdistance histogram.
Notes
The algorithm runs in \(O(E)\) time.
If enabled during compilation, this algorithm runs in parallel.
Examples
>>> from numpy import arange >>> from numpy.random import random >>> g = gt.random_graph(1000, lambda: (5, 5)) >>> eprop = g.new_edge_property("double") >>> eprop.get_array()[:] = random(g.num_edges()) >>> print(gt.edge_average(g, eprop)) (0.49888156584192045, 0.004096739923418754)

graph_tool.stats.
remove_labeled_edges
(g, label)[source]¶ Remove every edge e such that label[e] != 0.

graph_tool.stats.
label_parallel_edges
(g, mark_only=False, eprop=None)[source]¶ Label edges which are parallel, i.e, have the same source and target vertices. For each parallel edge set \(PE\), the labelling starts from 0 to \(PE1\). If mark_only==True, all parallel edges are simply marked with the value 1. If the eprop parameter is given (a
PropertyMap
), the labelling is stored there.

graph_tool.stats.
remove_parallel_edges
(g)[source]¶ Remove all parallel edges from the graph. Only one edge from each parallel edge set is left.

graph_tool.stats.
label_self_loops
(g, mark_only=False, eprop=None)[source]¶ Label edges which are selfloops, i.e, the source and target vertices are the same. For each selfloop edge set \(SL\), the labelling starts from 0 to \(SL1\). If mark_only == True, selfloops are labeled with 1 and others with 0. If the eprop parameter is given (a
PropertyMap
), the labelling is stored there.

graph_tool.stats.
distance_histogram
(g, weight=None, bins=[0, 1], samples=None, float_count=True)[source]¶ Return the shortestdistance histogram for each vertex pair in the graph.
Parameters: g :
Graph
Graph to be used.
weight :
PropertyMap
(optional, default: None)Edge weights.
bins : list of bins (optional, default: [0, 1])
List of bins to be used for the histogram. The values given represent the edges of the bins (i.e. lower and upper bounds). If the list contains two values, this will be used to automatically create an appropriate bin range, with a constant width given by the second value, and starting from the first value.
samples : int (optional, default: None)
If supplied, the distances will be randomly sampled from a number of source vertices given by this parameter. It samples is None (default), all pairs are used.
float_count : bool (optional, default: True)
If True, the counts in each histogram bin will be returned as floats. If False, they will be returned as integers.
Returns: counts :
ndarray
The bin counts.
bins :
ndarray
The bin edges.
See also
vertex_hist
 Vertex histograms.
edge_hist
 Edge histograms.
vertex_average
 Average of vertex degree, properties.
distance_histogram
 Shortestdistance histogram.
Notes
The algorithm runs in \(O(V^2)\) time, or \(O(V^2\log V)\) if weight is not None. If samples is supplied, the complexities are \(O(\text{samples}\times V)\) and \(O(\text{samples}\times V\log V)\), respectively.
If enabled during compilation, this algorithm runs in parallel.
Examples
>>> g = gt.random_graph(100, lambda: (3, 3)) >>> hist = gt.distance_histogram(g) >>> print(hist) [array([ 0., 300., 880., 2269., 3974., 2358., 119.]), array([0, 1, 2, 3, 4, 5, 6, 7], dtype=uint64)] >>> hist = gt.distance_histogram(g, samples=10) >>> print(hist) [array([ 0., 30., 87., 223., 394., 239., 17.]), array([0, 1, 2, 3, 4, 5, 6, 7], dtype=uint64)]