distances

Shortest path distances of all node pairs

Syntax

d = distances(G)

d = distances(G,s)

d = distances(G,s,t)

d = distances(___,'Method',algorithm)

Description

d = distances(G) returns a matrix, d, where d(i,j) is the length of the shortest path between node i and node j. If the graph is weighted (that is, G.Edges contains a variable Weight), then those weights are used as the distances along the edges in the graph. Otherwise, all edge distances are taken to be 1.

d = distances(G,s) restricts the source nodes to the nodes defined by s, such that d(i,j) is the distance from node s(i) to node j.

d = distances(G,s,t) additionally restricts the target nodes to the nodes defined by t, such that d(i,j) is the distance from node s(i) to node t(j).

d = distances(___,'Method',algorithm) optionally specifies the algorithm to use in computing the shortest path using any of the input arguments in previous syntaxes. For example, if G is a weighted graph, then distances(G,'Method','unweighted') ignores the edge weights in G and instead treats all edge weights as 1.

Examples

Shortest Path Distance for All Node Pairs

Create and plot a graph.

s = [1 1 1 2 5 5 5 8 9];
t = [2 3 4 5 6 7 8 9 10];
G = graph(s,t);
plot(G)

Calculate the shortest path distance between all node pairs in the graph. Since the graph edges do not have weights, all edge distances are taken to be 1.

d = distances(G)

d = 10×10

     0     1     1     1     2     3     3     3     4     5
     1     0     2     2     1     2     2     2     3     4
     1     2     0     2     3     4     4     4     5     6
     1     2     2     0     3     4     4     4     5     6
     2     1     3     3     0     1     1     1     2     3
     3     2     4     4     1     0     2     2     3     4
     3     2     4     4     1     2     0     2     3     4
     3     2     4     4     1     2     2     0     1     2
     4     3     5     5     2     3     3     1     0     1
     5     4     6     6     3     4     4     2     1     0

d is symmetric because G is an undirected graph. In general d(i,j) is the length of the shortest path between node i and node j, and for undirected graphs this is equivalent to d(j,i).

For example, find the length of the shortest path between node 1 and node 10.

d(1,10)

ans = 
5

Shortest Path Distances from Specified Sources

Create and plot a graph.

s = [1 1 1 1 2 2 3 4 4 5 6];
t = [2 3 4 5 3 6 6 5 7 7 7];
G = graph(s,t);
plot(G)

Find the shortest path distances from node 1, node 2, and node 3 to all other nodes in the graph.

d = distances(G,[1 2 3])

d = 3×7

     0     1     1     1     1     2     2
     1     0     1     2     2     1     2
     1     1     0     2     2     1     2

Use d to find the shortest path distance from node 1 to node 7.

d(1,7)

ans = 
2

Shortest Path Distances to Specified Targets

Create and plot a graph.

s = [1 1 1 2 2 3 3 4 5 5 6 7 8  8 10 11];
t = [2 3 10 4 12 5 4 6 6 7 9 8 9 11 11 12];
G = graph(s,t);
plot(G)

Find the shortest path distances from nodes 5 and 7 to nodes 2 and 3.

sources = [5 7];
targets = [2 3];
d = distances(G,sources,targets)

d = 2×2

     3     1
     4     2

Use d to find the shortest path distance between node 7 and node 3. In this case, d(i,j) is the distance from node sources(i) to node targets(j).

d(2,2)

ans = 
2

Ignore Edge Weights

Create and plot a directed graph with weighted edges.

s = [1 1 1 2 5 3 6 4 7 8 8 8];
t = [2 3 4 5 3 6 4 7 2 6 7 5];
weights = [100 10 10 10 10 20 10 30 50 10 70 10];
G = digraph(s,t,weights);
plot(G,'EdgeLabel',G.Edges.Weight)

Find the shortest path distance between all pairs of graph nodes.

d = distances(G)

d = 8×8

     0    90    10    10   100    30    40   Inf
   Inf     0    20    50    10    40    80   Inf
   Inf   110     0    30   120    20    60   Inf
   Inf    80   100     0    90   120    30   Inf
   Inf   120    10    40     0    30    70   Inf
   Inf    90   110    10   100     0    40   Inf
   Inf    50    70   100    60    90     0   Inf
   Inf   100    20    20    10    10    50     0

Since G is a directed graph, d is not symmetric, and d(i,j) corresponds to the distance between nodes i and j. The Inf values in d correspond to nodes that are unreachable. For example, since node 1 has no predecessors, it is not possible to reach node 1 from any other node in the graph. So the first column of d contains many Inf values to reflect that node 1 is unreachable.

By default, distances uses the edge weights to compute the distances. Specify 'Method' as 'unweighted' to ignore the edge weights and treat all edge distances as 1.

d1 = distances(G,'Method','unweighted')

d1 = 8×8

     0     1     1     1     2     2     2   Inf
   Inf     0     2     4     1     3     5   Inf
   Inf     4     0     2     5     1     3   Inf
   Inf     2     4     0     3     5     1   Inf
   Inf     5     1     3     0     2     4   Inf
   Inf     3     5     1     4     0     2   Inf
   Inf     1     3     5     2     4     0   Inf
   Inf     2     2     2     1     1     1     0

Input Arguments

G — Input graph

graph object | digraph object

Input graph, specified as either a graph or digraph object. Use graph to create an undirected graph or digraph to create a directed graph.

Example: G = graph(1,2)

Example: G = digraph([1 2],[2 3])

s — Source nodes

'all' (default) | node indices | node names

Source nodes, specified as one or more node indices or node names, or 'all' to select all source nodes.

This table shows the different ways to refer to one or more nodes either by their numeric node indices or by their node names.

Form	Single Node	Multiple Nodes
Node index	Scalar Example: 1	Vector Example: [1 2 3]
Node name	Character vector Example: 'A'	Cell array of character vectors Example: {'A' 'B' 'C'}
	String scalar Example: "A"	String array Example: ["A" "B" "C"]

s and t must not specify nodes named 'all' or 'Method', since these node names conflict with option names. Use findnode to instead pass in the node index for these cases.

Example: distances(G,[1 2])

Example: distances(G,'all',[1 3 5])

t — Target nodes

'all' (default) | node indices | node names

Target nodes, specified as one or more node indices or node names, or 'all' to select all target nodes.

s and t must not specify nodes named 'all' or 'Method', since these node names conflict with option names. Use findnode to instead pass in the node index for these cases.

Example: distances(G,[1 2])

Example: distances(G,'all',[1 3 5])

algorithm — Shortest path algorithm

'auto' (default) | 'unweighted' | 'positive' | 'mixed'

Shortest path algorithm, specified as one of the options in the table.

Option	Description
'auto' (default)	The 'auto' option automatically selects the algorithm: 'unweighted' is used for graph and digraph inputs with no edge weights. 'positive' is used for all graph inputs that have edge weights, and requires the weights to be nonnegative. This option is also used for digraph inputs with nonnegative edge weights. 'mixed' is used for digraph inputs whose edge weights contain some negative values. The graph cannot have negative cycles.
'unweighted'	Breadth-First computation that treats all edge weights as 1.
'positive'	Dijkstra algorithm that requires all edge weights to be nonnegative.
'mixed' (only for digraph)	Bellman-Ford algorithm for directed graphs that requires the graph to have no negative cycles. While 'mixed' is slower than 'positive' for the same problem, 'mixed' is more versatile as it allows some edge weights to be negative.

Note

For most graphs, 'unweighted' is the fastest algorithm, followed by 'positive', and 'mixed'.

Example: distances(G,s,t,'Method','unweighted')

Output Arguments

d — Shortest path distances between node pairs

matrix

Shortest path distances between node pairs, returned as a matrix. The size of d is (# source nodes)-by-(# target nodes). A value of Inf indicates a path that does not exist.

Tips

• The shortestpath, shortestpathtree, and distances functions do not support undirected graphs with negative edge weights, or more generally any graph containing a negative cycle, for these reasons:

  • A negative cycle is a path that leads from a node back to itself, with the sum of the edge weights on the path being negative. If a negative cycle is on a path between two nodes, then no shortest path exists between the nodes, since a shorter path can always be found by traversing the negative cycle.

  • A single negative edge weight in an undirected graph creates a negative cycle.

Extended Capabilities

Thread-Based Environment
Run code in the background using MATLAB® backgroundPool or accelerate code with Parallel Computing Toolbox™ ThreadPool.

Version History

Introduced in R2015b