Common graphs#
All graphs in Sage can be built through the graphs object. In order to
build a complete graph on 15 elements, one can do:
sage: g = graphs.CompleteGraph(15)
To get a path with 4 vertices, and the house graph:
sage: p = graphs.PathGraph(4)
sage: h = graphs.HouseGraph()
More interestingly, one can get the list of all graphs that Sage knows how to
build by typing graphs. in Sage and then hitting Tab.
Basic structures
Small Graphs
A small graph is just a single graph and has no parameter influencing the number of edges or vertices.
Platonic solids (ordered ascending by number of vertices)
Families of graphs
A family of graph is an infinite set of graphs which can be indexed by fixed number of parameters, e.g. two integer parameters. (A method whose name starts with a small letter does not return a single graph object but a graph iterator or a list of graphs or …)
Graphs from classical geometries over finite fields
A number of classes of graphs related to geometries over finite fields and quadrics and Hermitean varieties there.
Chessboard Graphs
Intersection graphs
These graphs are generated by geometric representations. The objects of the representation correspond to the graph vertices and the intersections of objects yield the graph edges.
Random graphs
Graphs with a given degree sequence
Miscellaneous
AUTHORS:
Robert Miller (2006-11-05): initial version, empty, random, petersen
Emily Kirkman (2006-11-12): basic structures, node positioning for all constructors
Emily Kirkman (2006-11-19): docstrings, examples
William Stein (2006-12-05): Editing.
Robert Miller (2007-01-16): Cube generation and plotting
Emily Kirkman (2007-01-16): more basic structures, docstrings
Emily Kirkman (2007-02-14): added more named graphs
Robert Miller (2007-06-08-11): Platonic solids, random graphs, graphs with a given degree sequence, random directed graphs
Robert Miller (2007-10-24): Isomorph free exhaustive generation
Nathann Cohen (2009-08-12): WorldMap
Michael Yurko (2009-9-01): added hyperstar, (n,k)-star, n-star, and bubblesort graphs
Anders Jonsson (2009-10-15): added generalized Petersen graphs
Harald Schilly and Yann Laigle-Chapuy (2010-03-24): added Fibonacci Tree
Jason Grout (2010-06-04): cospectral_graphs
Edward Scheinerman (2010-08-11): RandomTree
Ed Scheinerman (2010-08-21): added Grotzsch graph and Mycielski graphs
Ed Scheinerman (2010-11-15): added RandomTriangulation
Minh Van Nguyen (2010-11-26): added more named graphs
Keshav Kini (2011-02-16): added Shrikhande and Dyck graphs
David Coudert (2012-02-10): new RandomGNP generator
David Coudert (2012-08-02): added chessboard graphs: Queen, King, Knight, Bishop, and Rook graphs
Nico Van Cleemput (2013-05-26): added fullerenes
Nico Van Cleemput (2013-07-01): added benzenoids
Birk Eisermann (2013-07-29): new section ‘intersection graphs’, added (random, bounded) tolerance graphs
Marco Cognetta (2016-03-03): added TuranGraph
Functions and methods#
- class sage.graphs.graph_generators.GraphGenerators#
Bases:
objectA class consisting of constructors for several common graphs, as well as orderly generation of isomorphism class representatives. See the
module's helpfor a list of supported constructors.A list of all graphs and graph structures (other than isomorphism class representatives) in this database is available via tab completion. Type “graphs.” and then hit the Tab key to see which graphs are available.
The docstrings include educational information about each named graph with the hopes that this class can be used as a reference.
For all the constructors in this class (except the octahedral, dodecahedral, random and empty graphs), the position dictionary is filled to override the spring-layout algorithm.
ORDERLY GENERATION:
graphs(vertices, property=lambda x: True, augment='edges', size=None)This syntax accesses the generator of isomorphism class representatives. Iterates over distinct, exhaustive representatives.
Also: see the use of the nauty package for generating graphs at the
nauty_geng()method.INPUT:
vertices– a natural number orNoneto infinitely generate bigger and bigger graphs.property– (default:lambda x: True) any property to be tested on graphs before generation, but note that in general the graphs produced are not the same as those produced by using the property function to filter a list of graphs produced by using thelambda x: Truedefault. The generation process assumes the property has certain characteristics set by theaugmentargument, and only in the case of inherited properties such that all subgraphs of the relevant kind (foraugment='edges'oraugment='vertices') of a graph with the property also possess the property will there be no missing graphs. (Thepropertyargument is ignored ifdegree_sequenceis specified.)augment– (default:'edges') possible values:'edges'– augments a fixed number of vertices by adding one edge. In this case, all graphs on exactlyn=verticesare generated. If for any graph G satisfying the property, every subgraph, obtained from G by deleting one edge but not the vertices incident to that edge, satisfies the property, then this will generate all graphs with that property. If this does not hold, then all the graphs generated will satisfy the property, but there will be some missing.'vertices'– augments by adding a vertex and edges incident to that vertex. In this case, all graphs up ton=verticesare generated. If for any graph G satisfying the property, every subgraph, obtained from G by deleting one vertex and only edges incident to that vertex, satisfies the property, then this will generate all graphs with that property. If this does not hold, then all the graphs generated will satisfy the property, but there will be some missing.
size– (default:None) the size of the graph to be generated.degree_sequence– (default:None) a sequence of non-negative integers, orNone. If specified, the generated graphs will have these integers for degrees. In this case, property and size are both ignored.loops– (default:False) whether to allow loops in the graph or not.sparse– (default:True); whether to use a sparse or dense data structure. See the documentation ofGraph.copy(boolean) – If set toTrue(default) this method makes copies of the graphs before returning them. If set toFalsethe method returns the graph it is working on. The second alternative is faster, but modifying any of the graph instances returned by the method may break the function’s behaviour, as it is using these graphs to compute the next ones: only usecopy = Falsewhen you stick to reading the graphs returned.
EXAMPLES:
Print graphs on 3 or less vertices:
sage: for G in graphs(3, augment='vertices'): ....: print(G) Graph on 0 vertices Graph on 1 vertex Graph on 2 vertices Graph on 3 vertices Graph on 3 vertices Graph on 3 vertices Graph on 2 vertices Graph on 3 vertices
Print graphs on 3 vertices.
sage: for G in graphs(3): ....: print(G) Graph on 3 vertices Graph on 3 vertices Graph on 3 vertices Graph on 3 vertices
Generate all graphs with 5 vertices and 4 edges.
sage: L = graphs(5, size=4) sage: len(list(L)) 6
Generate all graphs with 5 vertices and up to 4 edges.
sage: L = list(graphs(5, lambda G: G.size() <= 4)) sage: len(L) 14 sage: graphs_list.show_graphs(L) # long time # needs sage.plot
Generate all graphs with up to 5 vertices and up to 4 edges.
sage: L = list(graphs(5, lambda G: G.size() <= 4, augment='vertices')) sage: len(L) 31 sage: graphs_list.show_graphs(L) # long time # needs sage.plot
Generate all graphs with degree at most 2, up to 6 vertices.
sage: property = lambda G: ( max([G.degree(v) for v in G] + [0]) <= 2 ) sage: L = list(graphs(6, property, augment='vertices')) sage: len(L) 45
Generate all bipartite graphs on up to 7 vertices: (see OEIS sequence A033995)
sage: L = list( graphs(7, lambda G: G.is_bipartite(), augment='vertices') ) sage: [len([g for g in L if g.order() == i]) for i in [1..7]] [1, 2, 3, 7, 13, 35, 88]
Generate all bipartite graphs on exactly 7 vertices:
sage: L = list( graphs(7, lambda G: G.is_bipartite()) ) sage: len(L) 88
Generate all bipartite graphs on exactly 8 vertices:
sage: L = list( graphs(8, lambda G: G.is_bipartite()) ) # long time sage: len(L) # long time 303
Remember that the property argument does not behave as a filter, except for appropriately inheritable properties:
sage: property = lambda G: G.is_vertex_transitive() sage: len(list(graphs(4, property))) # needs sage.groups 1 sage: sum(1 for g in graphs(4) if property(g)) # needs sage.groups 4 sage: property = lambda G: G.is_bipartite() sage: len(list(graphs(4, property))) 7 sage: sum(1 for g in graphs(4) if property(g)) 7
Generate graphs on the fly: (see OEIS sequence A000088)
sage: for i in range(7): ....: print(len(list(graphs(i)))) 1 1 2 4 11 34 156
Generate all simple graphs, allowing loops: (see OEIS sequence A000666)
sage: L = list(graphs(5,augment='vertices',loops=True)) # long time sage: for i in [0..5]: # long time ....: print((i, len([g for g in L if g.order() == i]))) (0, 1) (1, 2) (2, 6) (3, 20) (4, 90) (5, 544)
Generate all graphs with a specified degree sequence (see OEIS sequence A002851):
sage: for i in [4,6,8]: # long time (4s on sage.math, 2012) ....: print((i, len([g for g in graphs(i, degree_sequence=[3]*i) if g.is_connected()]))) (4, 1) (6, 2) (8, 5) sage: for i in [4,6,8]: # long time (7s on sage.math, 2012) ....: print((i, len([g for g in graphs(i, augment='vertices', degree_sequence=[3]*i) if g.is_connected()]))) (4, 1) (6, 2) (8, 5)
sage: print((10, len([g for g in graphs(10,degree_sequence=[3]*10) if g.is_connected()]))) # not tested (10, 19)
Make sure that the graphs are really independent and the generator survives repeated vertex removal (github issue #8458):
sage: for G in graphs(3): ....: G.delete_vertex(0) ....: print(G.order()) 2 2 2 2
REFERENCE:
Brendan D. McKay, Isomorph-Free Exhaustive generation. Journal of Algorithms, Volume 26, Issue 2, February 1998, pages 306-324.
- static AffineOrthogonalPolarGraph(d, q, sign='+')#
Return the affine polar graph \(VO^+(d,q),VO^-(d,q)\) or \(VO(d,q)\).
Affine Polar graphs are built from a \(d\)-dimensional vector space over \(F_q\), and a quadratic form which is hyperbolic, elliptic or parabolic according to the value of
sign.Note that \(VO^+(d,q),VO^-(d,q)\) are strongly regular graphs, while \(VO(d,q)\) is not.
For more information on Affine Polar graphs, see Affine Polar Graphs page of Andries Brouwer’s website.
INPUT:
d– integer;dmust be even ifsign is not None, and odd otherwiseq– integer; a power of a prime number, as \(F_q\) must existsign– string (default:"+"); must be equal to"+","-", orNoneto compute (respectively) \(VO^+(d,q),VO^-(d,q)\) or \(VO(d,q)\)
Note
The graph \(VO^\epsilon(d,q)\) is the graph induced by the non-neighbors of a vertex in an
Orthogonal Polar Graph\(O^\epsilon(d+2,q)\).EXAMPLES:
The
Brouwer-Haemers graphis isomorphic to \(VO^-(4,3)\):sage: g = graphs.AffineOrthogonalPolarGraph(4,3,"-") # needs sage.libs.gap sage: g.is_isomorphic(graphs.BrouwerHaemersGraph()) # needs sage.libs.gap True
Some examples from Brouwer’s table or strongly regular graphs:
sage: # needs sage.libs.gap sage: g = graphs.AffineOrthogonalPolarGraph(6,2,"-"); g Affine Polar Graph VO^-(6,2): Graph on 64 vertices sage: g.is_strongly_regular(parameters=True) (64, 27, 10, 12) sage: g = graphs.AffineOrthogonalPolarGraph(6,2,"+"); g Affine Polar Graph VO^+(6,2): Graph on 64 vertices sage: g.is_strongly_regular(parameters=True) (64, 35, 18, 20)
When
sign is None:sage: # needs sage.libs.gap sage: g = graphs.AffineOrthogonalPolarGraph(5,2,None); g Affine Polar Graph VO^-(5,2): Graph on 32 vertices sage: g.is_strongly_regular(parameters=True) False sage: g.is_regular() True sage: g.is_vertex_transitive() True
- static AfricaMap(continental=False, year=2018)#
Return African states as a graph of common border.
“African state” here is defined as an independent state having the capital city in Africa. The graph has an edge between those countries that have common land border.
INPUT:
continental– boolean (default:False); whether to only return states in the continental Africa or all African statesyear– integer (default:2018); reserved for future use
EXAMPLES:
sage: Africa = graphs.AfricaMap(); Africa Africa Map: Graph on 54 vertices sage: sorted(Africa.neighbors('Libya')) ['Algeria', 'Chad', 'Egypt', 'Niger', 'Sudan', 'Tunisia'] sage: cont_Africa = graphs.AfricaMap(continental=True) sage: cont_Africa.order() 48 sage: 'Madagaskar' in cont_Africa False
- static AhrensSzekeresGeneralizedQuadrangleGraph(q, dual=False)#
Return the collinearity graph of the generalized quadrangle \(AS(q)\), or of its dual
Let \(q\) be an odd prime power. \(AS(q)\) is a generalized quadrangle (Wikipedia article Generalized_quadrangle) of order \((q-1,q+1)\), see 3.1.5 in [PT2009]. Its points are elements of \(F_q^3\), and lines are sets of size \(q\) of the form
\(\{ (\sigma, a, b) \mid \sigma\in F_q \}\)
\(\{ (a, \sigma, b) \mid \sigma\in F_q \}\)
\(\{ (c \sigma^2 - b \sigma + a, -2 c \sigma + b, \sigma) \mid \sigma\in F_q \}\),
where \(a\), \(b\), \(c\) are arbitrary elements of \(F_q\).
INPUT:
q– a power of an odd prime numberdual– boolean (default:False); whether to return the collinearity graph of \(AS(q)\) or of the dual \(AS(q)\) (whenTrue)
EXAMPLES:
sage: g = graphs.AhrensSzekeresGeneralizedQuadrangleGraph(5); g AS(5); GQ(4, 6): Graph on 125 vertices sage: g.is_strongly_regular(parameters=True) (125, 28, 3, 7) sage: g = graphs.AhrensSzekeresGeneralizedQuadrangleGraph(5, dual=True); g AS(5)*; GQ(6, 4): Graph on 175 vertices sage: g.is_strongly_regular(parameters=True) (175, 30, 5, 5)
- static AlternatingFormsGraph(n, q)#
Return the alternating forms graph with the given parameters.
This builds a graph whose vertices are all \(n`x`n\) skew-symmetric matrices over \(GF(q)\) with zero diagonal. Two vertices are adjacent if and only if the difference of the two matrices has rank 2.
This grap is distance-regular with classical parameters \((\lfloor \frac n 2 \rfloor, q^2, q^2 - 1, q^{2 \lceil \frac n 2 \rceil -1})\).
INPUT:
n– integerq– a prime power
EXAMPLES:
sage: G = graphs.AlternatingFormsGraph(5, 2) # long time sage: G.is_distance_regular(True) # long time ([155, 112, None], [None, 1, 20])
REFERENCES:
See [BCN1989] pp. 282-284 for a rather detailed discussion, otherwise see [VDKT2016] p. 22.
- static AztecDiamondGraph(n)#
Return the Aztec Diamond graph of order
n.See the Wikipedia article Aztec_diamond for more information.
EXAMPLES:
sage: graphs.AztecDiamondGraph(2) Aztec Diamond graph of order 2 sage: [graphs.AztecDiamondGraph(i).num_verts() for i in range(8)] [0, 4, 12, 24, 40, 60, 84, 112] sage: [graphs.AztecDiamondGraph(i).num_edges() for i in range(8)] [0, 4, 16, 36, 64, 100, 144, 196] sage: G = graphs.AztecDiamondGraph(3) sage: sum(1 for p in G.perfect_matchings()) 64
- static Balaban10Cage(embedding=1)#
Return the Balaban 10-cage.
The Balaban 10-cage is a 3-regular graph with 70 vertices and 105 edges. See the Wikipedia article Balaban_10-cage.
The default embedding gives a deeper understanding of the graph’s automorphism group. It is divided into 4 layers (each layer being a set of points at equal distance from the drawing’s center). From outside to inside:
L1: The outer layer (vertices which are the furthest from the origin) is actually the disjoint union of two cycles of length 10.
L2: The second layer is an independent set of 20 vertices.
L3: The third layer is a matching on 10 vertices.
L4: The inner layer (vertices which are the closest from the origin) is also the disjoint union of two cycles of length 10.
This graph is not vertex-transitive, and its vertices are partitioned into 3 orbits: L2, L3, and the union of L1 of L4 whose elements are equivalent.
INPUT:
embedding– integer (default:1); two embeddings are available, and can be selected by settingembeddingto be either 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.Balaban10Cage() sage: g.girth() 10 sage: g.chromatic_number() 2 sage: g.diameter() 6 sage: g.is_hamiltonian() # needs sage.numerical.mip True sage: g.show(figsize=[10,10]) # long time # needs sage.plot
- static Balaban11Cage(embedding=1)#
Return the Balaban 11-cage.
For more information, see the Wikipedia article Balaban_11-cage.
INPUT:
embedding– integer (default:1); three embeddings are available, and can be selected by settingembeddingto be 1, 2, or 3The first embedding is the one appearing on page 9 of the Fifth Annual Graph Drawing Contest report [EMMN1998]. It separates vertices based on their eccentricity (see
eccentricity()).The second embedding has been produced just for Sage and is meant to emphasize the automorphism group’s 6 orbits.
The last embedding is the default one produced by the
LCFGraph()constructor.
Note
The vertex labeling changes according to the value of
embedding=1.EXAMPLES:
Basic properties:
sage: g = graphs.Balaban11Cage() sage: g.order() 112 sage: g.size() 168 sage: g.girth() 11 sage: g.diameter() 8 sage: g.automorphism_group().cardinality() # needs sage.groups 64
Our many embeddings:
sage: g1 = graphs.Balaban11Cage(embedding=1) sage: g2 = graphs.Balaban11Cage(embedding=2) # needs networkx sage: g3 = graphs.Balaban11Cage(embedding=3) # needs networkx sage: g1.show(figsize=[10,10]) # long time # needs sage.plot sage: g2.show(figsize=[10,10]) # long time # needs networkx sage.plot sage: g3.show(figsize=[10,10]) # long time # needs networkx sage.plot
Proof that the embeddings are the same graph:
sage: g1.is_isomorphic(g2) # g2 and g3 are obviously isomorphic # needs networkx True
- static BalancedTree(r, h)#
Returns the perfectly balanced tree of height \(h \geq 1\), whose root has degree \(r \geq 2\).
The number of vertices of this graph is \(1 + r + r^2 + \cdots + r^h\), that is, \(\frac{r^{h+1} - 1}{r - 1}\). The number of edges is one less than the number of vertices.
INPUT:
r– positive integer \(\geq 2\). The degree of the root node.h– positive integer \(\geq 1\). The height of the balanced tree.
OUTPUT:
The perfectly balanced tree of height \(h \geq 1\) and whose root has degree \(r \geq 2\). A
NetworkXErroris returned if \(r < 2\) or \(h < 1\).ALGORITHM:
Uses NetworkX.
EXAMPLES:
A balanced tree whose root node has degree \(r = 2\), and of height \(h = 1\), has order 3 and size 2:
sage: G = graphs.BalancedTree(2, 1); G # needs networkx Balanced tree: Graph on 3 vertices sage: G.order(); G.size() # needs networkx 3 2 sage: r = 2; h = 1 sage: v = 1 + r sage: v; v - 1 3 2
Plot a balanced tree of height 5, whose root node has degree \(r = 3\):
sage: G = graphs.BalancedTree(3, 5) # needs networkx sage: G.show() # long time # needs networkx sage.plot
A tree is bipartite. If its vertex set is finite, then it is planar.
sage: # needs networkx sage: r = randint(2, 5); h = randint(1, 7) sage: T = graphs.BalancedTree(r, h) sage: T.is_bipartite() True sage: T.is_planar() True sage: v = (r^(h + 1) - 1) / (r - 1) sage: T.order() == v True sage: T.size() == v - 1 True
- static BarbellGraph(n1, n2)#
Returns a barbell graph with
2*n1 + n2nodes. The argumentn1must be greater than or equal to 2.A barbell graph is a basic structure that consists of a path graph of order
n2connecting two complete graphs of ordern1each.INPUT:
n1– integer \(\geq 2\). The order of each of the two complete graphs.n2– nonnegative integer. The order of the path graph connecting the two complete graphs.
OUTPUT:
A barbell graph of order
2*n1 + n2. AValueErroris returned ifn1 < 2orn2 < 0.PLOTTING:
Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each barbell graph will be displayed with the two complete graphs in the lower-left and upper-right corners, with the path graph connecting diagonally between the two. Thus the
n1-th node will be drawn at a 45 degree angle from the horizontal right center of the first complete graph, and then1 + n2 + 1-th node will be drawn 45 degrees below the left horizontal center of the second complete graph.EXAMPLES:
Construct and show a barbell graph
Bar = 4,Bells = 9:sage: g = graphs.BarbellGraph(9, 4); g Barbell graph: Graph on 22 vertices sage: g.show() # long time # needs sage.plot
An
n1 >= 2,n2 >= 0barbell graph has order2*n1 + n2. It has the complete graph onn1vertices as a subgraph. It also has the path graph onn2vertices as a subgraph.sage: n1 = randint(2, 2*10^2) sage: n2 = randint(0, 2*10^2) sage: g = graphs.BarbellGraph(n1, n2) sage: v = 2*n1 + n2 sage: g.order() == v True sage: K_n1 = graphs.CompleteGraph(n1) sage: P_n2 = graphs.PathGraph(n2) sage: # needs sage.modules sage: s_K = g.subgraph_search(K_n1, induced=True) sage: s_P = g.subgraph_search(P_n2, induced=True) sage: K_n1.is_isomorphic(s_K) True sage: P_n2.is_isomorphic(s_P) True
- static BidiakisCube()#
Return the Bidiakis cube.
For more information, see the Wikipedia article Bidiakis_cube.
EXAMPLES:
The Bidiakis cube is a 3-regular graph having 12 vertices and 18 edges. This means that each vertex has a degree of 3:
sage: g = graphs.BidiakisCube(); g Bidiakis cube: Graph on 12 vertices sage: g.show() # long time # needs sage.plot sage: g.order() 12 sage: g.size() 18 sage: g.is_regular(3) True
It is a Hamiltonian graph with diameter 3 and girth 4:
sage: g.is_hamiltonian() # needs sage.numerical.mip True sage: g.diameter() 3 sage: g.girth() 4
It is a planar graph with characteristic polynomial \((x - 3) (x - 2) (x^4) (x + 1) (x + 2) (x^2 + x - 4)^2\) and chromatic number 3:
sage: g.is_planar() True sage: char_poly = g.characteristic_polynomial() # needs sage.modules sage: x = char_poly.parent()('x') # needs sage.modules sage: char_poly == (x - 3) * (x - 2) * (x^4) * (x + 1) * (x + 2) * (x^2 + x - 4)^2 # needs sage.modules True sage: g.chromatic_number() # needs sage.modules 3
- static BiggsSmithGraph(embedding=1)#
Return the Biggs-Smith graph.
For more information, see the Wikipedia article Biggs-Smith_graph.
INPUT:
embedding– integer (default:1); two embeddings are available, and can be selected by settingembeddingto be 1 or 2
EXAMPLES:
Basic properties:
sage: # needs networkx sage: g = graphs.BiggsSmithGraph() sage: g.order() 102 sage: g.size() 153 sage: g.girth() 9 sage: g.diameter() 7 sage: g.automorphism_group().cardinality() # long time 2448 sage: g.show(figsize=[10, 10]) # long time # needs sage.plot
The other embedding:
sage: graphs.BiggsSmithGraph(embedding=2).show() # long time # needs networkx
- static BilinearFormsGraph(d, e, q)#
Return a bilinear forms graph with the given parameters.
This builds a graph whose vertices are all \(d`x`e\) matrices over \(GF(q)\). Two vertices are adjacent if the difference of the two matrices has rank 1.
The graph is distance-regular with classical parameters \((\min(d, e), q, q-1 , q^{\max(d, e)}-1)\).
INPUT:
d, e– integers; dimension of the matricesq– integer; a prime power
EXAMPLES:
sage: # needs sage.modules sage: G = graphs.BilinearFormsGraph(3, 3, 2) sage: G.is_distance_regular(True) ([49, 36, 16, None], [None, 1, 6, 28]) sage: G = graphs.BilinearFormsGraph(3,3,3) # not tested (20 s) # needs sage.rings.finite_rings sage: G.order() # not tested (due to above) # needs sage.rings.finite_rings 19683 sage: G = graphs.BilinearFormsGraph(3, 4, 2) # long time # needs sage.rings.finite_rings sage: G.is_distance_regular(True) # long time # needs sage.rings.finite_rings ([105, 84, 48, None], [None, 1, 6, 28])
REFERENCES:
See [BCN1989] pp. 280-282 for a rather detailed discussion, otherwise see [VDKT2016] p. 21.
- static BishopGraph(dim_list, radius=None, relabel=False)#
Return the \(d\)-dimensional Bishop Graph with prescribed dimensions.
The 2-dimensional Bishop Graph of parameters \(n\) and \(m\) is a graph with \(nm\) vertices in which each vertex represents a square in an \(n \times m\) chessboard, and each edge corresponds to a legal move by a bishop.
The \(d\)-dimensional Bishop Graph with \(d >= 2\) has for vertex set the cells of a \(d\)-dimensional grid with prescribed dimensions, and each edge corresponds to a legal move by a bishop in any pairs of dimensions.
The Bishop Graph is not connected.
INPUT:
dim_list– iterable (list, set, dict); provides the dimensions \(n_1, n_2, \ldots, n_d\), with \(n_i \geq 1\), of the chessboardradius– integer (default:None); by setting the radius to a positive integer, one may decrease the power of the bishop to at mostradiussteps.relabel– boolean (default:False); indicates whether the vertices must be relabeled as integers
EXAMPLES:
The (n,m)-Bishop Graph is not connected:
sage: G = graphs.BishopGraph( [3, 4] ) sage: G.is_connected() False
The Bishop Graph can be obtained from Knight Graphs:
sage: for d in range(3,12): # long time ....: H = Graph() ....: for r in range(1,d+1): ....: B = graphs.BishopGraph([d,d],radius=r) ....: H.add_edges( graphs.KnightGraph([d,d],one=r,two=r).edges(sort=False) ) ....: if not B.is_isomorphic(H): ....: print("that's not good!")
- static BlanusaFirstSnarkGraph()#
Return the first Blanusa Snark Graph.
The Blanusa graphs are two snarks on 18 vertices and 27 edges. For more information on them, see the Wikipedia article Blanusa_snarks.
See also
EXAMPLES:
sage: g = graphs.BlanusaFirstSnarkGraph() sage: g.order() 18 sage: g.size() 27 sage: g.diameter() 4 sage: g.girth() 5 sage: g.automorphism_group().cardinality() # needs sage.groups 8
- static BlanusaSecondSnarkGraph()#
Return the second Blanusa Snark Graph.
The Blanusa graphs are two snarks on 18 vertices and 27 edges. For more information on them, see the Wikipedia article Blanusa_snarks.
See also
EXAMPLES:
sage: g = graphs.BlanusaSecondSnarkGraph() sage: g.order() 18 sage: g.size() 27 sage: g.diameter() 4 sage: g.girth() 5 sage: g.automorphism_group().cardinality() # needs sage.groups 4
- static BrinkmannGraph()#
Return the Brinkmann graph.
For more information, see the Wikipedia article Brinkmann_graph.
EXAMPLES:
The Brinkmann graph is a 4-regular graph having 21 vertices and 42 edges. This means that each vertex has degree 4:
sage: G = graphs.BrinkmannGraph(); G Brinkmann graph: Graph on 21 vertices sage: G.show() # long time # needs sage.plot sage: G.order() 21 sage: G.size() 42 sage: G.is_regular(4) True
It is an Eulerian graph with radius 3, diameter 3, and girth 5:
sage: G.is_eulerian() True sage: G.radius() 3 sage: G.diameter() 3 sage: G.girth() 5
The Brinkmann graph is also Hamiltonian with chromatic number 4:
sage: G.is_hamiltonian() # needs sage.numerical.mip True sage: G.chromatic_number() 4
Its automorphism group is isomorphic to \(D_7\):
sage: ag = G.automorphism_group() # needs sage.groups sage: ag.is_isomorphic(DihedralGroup(7)) # needs sage.groups True
- static BrouwerHaemersGraph()#
Return the Brouwer-Haemers Graph.
The Brouwer-Haemers is the only strongly regular graph of parameters \((81,20,1,6)\). It is build in Sage as the Affine Orthogonal graph \(VO^-(6,3)\). For more information on this graph, see its corresponding page on Andries Brouwer’s website.
EXAMPLES:
sage: g = graphs.BrouwerHaemersGraph(); g # needs sage.modules Brouwer-Haemers: Graph on 81 vertices
It is indeed strongly regular with parameters \((81,20,1,6)\):
sage: g.is_strongly_regular(parameters=True) # long time # needs sage.modules sage.rings.finite_rings (81, 20, 1, 6)
Its has as eigenvalues \(20,2\) and \(-7\):
sage: set(g.spectrum()) == {20,2,-7} # needs sage.modules sage.rings.finite_rings True
- static BubbleSortGraph(n)#
Returns the bubble sort graph \(B(n)\).
The vertices of the bubble sort graph are the set of permutations on \(n\) symbols. Two vertices are adjacent if one can be obtained from the other by swapping the labels in the \(i\)-th and \((i+1)\)-th positions for \(1 \leq i \leq n-1\). In total, \(B(n)\) has order \(n!\). Swapping two labels as described previously corresponds to multiplying on the right the permutation corresponding to the node by an elementary transposition in the
SymmetricGroup.The bubble sort graph is the underlying graph of the
permutahedron().INPUT:
n– positive integer. The number of symbols to permute.
OUTPUT:
The bubble sort graph \(B(n)\) on \(n\) symbols. If \(n < 1\), a
ValueErroris returned.EXAMPLES:
sage: g = graphs.BubbleSortGraph(4); g Bubble sort: Graph on 24 vertices sage: g.plot() # long time # needs sage.plot Graphics object consisting of 61 graphics primitives
The bubble sort graph on \(n = 1\) symbol is the trivial graph \(K_1\):
sage: graphs.BubbleSortGraph(1) Bubble sort: Graph on 1 vertex
If \(n \geq 1\), then the order of \(B(n)\) is \(n!\):
sage: n = randint(1, 8) sage: g = graphs.BubbleSortGraph(n) sage: g.order() == factorial(n) True
See also
AUTHORS:
Michael Yurko (2009-09-01)
- static BuckyBall()#
Return the Bucky Ball graph.
This graph is a 3-regular 60-vertex planar graph. Its vertices and edges correspond precisely to the carbon atoms and bonds in buckminsterfullerene. When embedded on a sphere, its 12 pentagon and 20 hexagon faces are arranged exactly as the sections of a soccer ball.
EXAMPLES:
The Bucky Ball is planar:
sage: g = graphs.BuckyBall() sage: g.is_planar() True
The Bucky Ball can also be created by extracting the 1-skeleton of the Bucky Ball polyhedron, but this is much slower:
sage: # needs sage.geometry.polyhedron sage.groups sage.rings.number_field sage: g = polytopes.buckyball().vertex_graph() sage: g.remove_loops() sage: h = graphs.BuckyBall() sage: g.is_isomorphic(h) True
The graph is returned along with an attractive embedding:
sage: g = graphs.BuckyBall() # long time sage: g.plot(vertex_labels=False, vertex_size=10).show() # long time, needs sage.plot
- static BullGraph()#
Return a bull graph with 5 nodes.
A bull graph is named for its shape. It’s a triangle with horns. See the Wikipedia article Bull_graph for more information.
PLOTTING:
Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the bull graph is drawn as a triangle with the first node (0) on the bottom. The second and third nodes (1 and 2) complete the triangle. Node 3 is the horn connected to 1 and node 4 is the horn connected to node 2.
EXAMPLES:
Construct and show a bull graph:
sage: g = graphs.BullGraph(); g Bull graph: Graph on 5 vertices sage: g.show() # long time # needs sage.plot
The bull graph has 5 vertices and 5 edges. Its radius is 2, its diameter 3, and its girth 3. The bull graph is planar with chromatic number 3 and chromatic index also 3:
sage: g.order(); g.size() 5 5 sage: g.radius(); g.diameter(); g.girth() 2 3 3 sage: g.chromatic_number() 3
The bull graph has chromatic polynomial \(x(x - 2)(x - 1)^3\) and Tutte polynomial \(x^4 + x^3 + x^2 y\). Its characteristic polynomial is \(x(x^2 - x - 3)(x^2 + x - 1)\), which follows from the definition of characteristic polynomials for graphs, i.e. \(\det(xI - A)\), where \(x\) is a variable, \(A\) the adjacency matrix of the graph, and \(I\) the identity matrix of the same dimensions as \(A\):
sage: # needs sage.libs.flint sage: chrompoly = g.chromatic_polynomial() sage: x = chrompoly.parent()('x') sage: x * (x - 2) * (x - 1)^3 == chrompoly True sage: # needs sage.libs.flint sage.modules sage: charpoly = g.characteristic_polynomial() sage: M = g.adjacency_matrix(); M [0 1 1 0 0] [1 0 1 1 0] [1 1 0 0 1] [0 1 0 0 0] [0 0 1 0 0] sage: Id = identity_matrix(ZZ, M.nrows()) sage: D = x*Id - M sage: D.determinant() == charpoly # needs sage.symbolic True sage: x * (x^2 - x - 3) * (x^2 + x - 1) == charpoly True
- static ButterflyGraph()#
Return the butterfly graph.
Let \(C_3\) be the cycle graph on 3 vertices. The butterfly or bowtie graph is obtained by joining two copies of \(C_3\) at a common vertex, resulting in a graph that is isomorphic to the friendship graph \(F_2\). See the Wikipedia article Butterfly_graph for more information.
See also
EXAMPLES:
The butterfly graph is a planar graph on 5 vertices and having 6 edges:
sage: G = graphs.ButterflyGraph(); G Butterfly graph: Graph on 5 vertices sage: G.show() # long time # needs sage.plot sage: G.is_planar() True sage: G.order() 5 sage: G.size() 6
It has diameter 2, girth 3, and radius 1:
sage: G.diameter() 2 sage: G.girth() 3 sage: G.radius() 1
The butterfly graph is Eulerian, with chromatic number 3:
sage: G.is_eulerian() True sage: G.chromatic_number() 3
- static CaiFurerImmermanGraph(G, twisted=False)#
Return the a Cai-Furer-Immerman graph from \(G\), possibly a twisted one, and a partition of its nodes.
A Cai-Furer-Immerman graph from/on \(G\) is a graph created by applying the transformation described in [CFI1992] on a graph \(G\), that is substituting every vertex v in \(G\) with a Furer gadget \(F(v)\) of order d equal to the degree of the vertex, and then substituting every edge \((v,u)\) in \(G\) with a pair of edges, one connecting the two “a” nodes of \(F(v)\) and \(F(u)\) and the other their two “b” nodes. The returned coloring of the vertices is made by the union of the colorings of each single Furer gadget, individualised for each vertex of \(G\). To understand better what these “a” and “b” nodes are, see the documentation on Furer gadgets.
Furthermore, this method can apply what is described in the paper mentioned above as a “twist” on an edge, that is taking only one of the pairs of edges introduced in the new graph and swap two of their extremes, making each edge go from an “a” node to a “b” node. This is only doable if the original graph G is connected.
A CaiFurerImmerman graph on a graph with no balanced vertex separators smaller than s and its twisted version cannot be distinguished by k-WL for any k < s.
INPUT:
G– An undirected graph on which to construct theCai-Furer-Immerman graph
twisted– A boolean indicating if the version to constructis a twisted one or not
OUTPUT:
H– The Cai-Furer-Immerman graph onGcoloring– A list of list of vertices, representing thepartition induced by the coloring on
H
EXAMPLES:
CaiFurerImmerman graph with no balanced vertex separator smaller than 2
sage: G = graphs.CycleGraph(4) sage: CFI, p = graphs.CaiFurerImmermanGraph(G) sage: sorted(CFI, key=str) [(0, ()), (0, (0, 'a')), (0, (0, 'b')), (0, (0, 1)), (0, (1, 'a')), (0, (1, 'b')), (1, ()), (1, (0, 'a')), (1, (0, 'b')), (1, (0, 1)), (1, (1, 'a')), (1, (1, 'b')), (2, ()), (2, (0, 'a')), (2, (0, 'b')), (2, (0, 1)), (2, (1, 'a')), (2, (1, 'b')), (3, ()), (3, (0, 'a')), (3, (0, 'b')), (3, (0, 1)), (3, (1, 'a')), (3, (1, 'b'))] sage: sorted(CFI.edge_iterator(), key=str) [((0, ()), (0, (0, 'b')), None), ((0, ()), (0, (1, 'b')), None), ((0, (0, 'a')), (1, (0, 'a')), None), ((0, (0, 'b')), (1, (0, 'b')), None), ((0, (0, 1)), (0, (0, 'a')), None), ((0, (0, 1)), (0, (1, 'a')), None), ((0, (1, 'a')), (3, (0, 'a')), None), ((0, (1, 'b')), (3, (0, 'b')), None), ((1, ()), (1, (0, 'b')), None), ((1, ()), (1, (1, 'b')), None), ((1, (0, 1)), (1, (0, 'a')), None), ((1, (0, 1)), (1, (1, 'a')), None), ((1, (1, 'a')), (2, (0, 'a')), None), ((1, (1, 'b')), (2, (0, 'b')), None), ((2, ()), (2, (0, 'b')), None), ((2, ()), (2, (1, 'b')), None), ((2, (0, 1)), (2, (0, 'a')), None), ((2, (0, 1)), (2, (1, 'a')), None), ((2, (1, 'a')), (3, (1, 'a')), None), ((2, (1, 'b')), (3, (1, 'b')), None), ((3, ()), (3, (0, 'b')), None), ((3, ()), (3, (1, 'b')), None), ((3, (0, 1)), (3, (0, 'a')), None), ((3, (0, 1)), (3, (1, 'a')), None)]
- static CameronGraph()#
Return the Cameron graph.
The Cameron graph is strongly regular with parameters \(v = 231, k = 30, \lambda = 9, \mu = 3\).
For more information on the Cameron graph, see https://www.win.tue.nl/~aeb/graphs/Cameron.html.
EXAMPLES:
sage: # needs sage.groups sage: g = graphs.CameronGraph() sage: g.order() 231 sage: g.size() 3465 sage: g.is_strongly_regular(parameters=True) # long time (231, 30, 9, 3)
- static Cell120()#
Return the 120-Cell graph.
This is the adjacency graph of the 120-cell. It has 600 vertices and 1200 edges. For more information, see the Wikipedia article 120-cell.
EXAMPLES:
sage: # long time, needs sage.rings.number_field sage: g = graphs.Cell120() sage: g.size() 1200 sage: g.is_regular(4) True sage: g.is_vertex_transitive() True
- static Cell600(embedding=1)#
Return the 600-Cell graph.
This is the adjacency graph of the 600-cell. It has 120 vertices and 720 edges. For more information, see the Wikipedia article 600-cell.
INPUT:
embedding– integer (default:1); two different embeddings for a plot
EXAMPLES:
sage: # long time, needs sage.rings.number_field sage: g = graphs.Cell600() sage: g.size() 720 sage: g.is_regular(12) True sage: g.is_vertex_transitive() True
- static ChessboardGraphGenerator(dim_list, rook=True, rook_radius=None, bishop=True, bishop_radius=None, knight=True, knight_x=1, knight_y=2, relabel=False)#
Return a Graph built on a \(d\)-dimensional chessboard with prescribed dimensions and interconnections.
This function allows to generate many kinds of graphs corresponding to legal movements on a \(d\)-dimensional chessboard: Queen Graph, King Graph, Knight Graphs, Bishop Graph, and many generalizations. It also allows to avoid redundant code.
INPUT:
dim_list– iterable (list, set, dict); provides the dimensions \(n_1, n_2, \ldots, n_d\), with \(n_i \geq 1\), of the chessboardrook– boolean (default:True); indicates whether the chess piece is able to move as a rook, that is at any distance along a dimensionrook_radius– integer (default:None); restriction on the rook-like movements to distance at mostrook_radiusbishop– boolean (default:True); indicates whether the chess piece is able to move like a bishop, that is along diagonalsbishop_radius– integer (default:None); restriction on the bishop-like movements to distance at mostbishop_radiusknight– boolean (default:True); indicating whether the chess piece is able to move like a knightknight_x– integer (default:1); indicates the number on steps the chess piece moves in one dimension when moving like a knightknight_y– integer (default:2); indicates the number on steps the chess piece moves in the second dimension when moving like a knightrelabel– boolean (default:False); indicates whether the vertices must be relabeled as integers
OUTPUT:
A Graph build on a \(d\)-dimensional chessboard with prescribed dimensions, and with edges according given parameters.
A string encoding the dimensions. This is mainly useful for providing names to graphs.
EXAMPLES:
A \((2,2)\)-King Graph is isomorphic to the complete graph on 4 vertices:
sage: G, _ = graphs.ChessboardGraphGenerator( [2,2] ) sage: G.is_isomorphic( graphs.CompleteGraph(4) ) True
A Rook’s Graph in 2 dimensions is isomorphic to the Cartesian product of 2 complete graphs:
sage: G, _ = graphs.ChessboardGraphGenerator([3,4], rook=True, rook_radius=None, bishop=False, knight=False) sage: H = (graphs.CompleteGraph(3)).cartesian_product(graphs.CompleteGraph(4)) sage: G.is_isomorphic(H) True
- static ChvatalGraph()#
Return the Chvatal graph.
Chvatal graph is one of the few known graphs to satisfy Grunbaum’s conjecture that for every \(m\), \(n\), there is an \(m\)-regular, \(m\)-chromatic graph of girth at least \(n\). For more information, see the Wikipedia article Chv%C3%A1tal_graph.
EXAMPLES:
The Chvatal graph has 12 vertices and 24 edges. It is a 4-regular, 4-chromatic graph with radius 2, diameter 2, and girth 4:
sage: G = graphs.ChvatalGraph(); G Chvatal graph: Graph on 12 vertices sage: G.order(); G.size() 12 24 sage: G.degree() [4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4] sage: G.chromatic_number() 4 sage: G.radius(); G.diameter(); G.girth() 2 2 4
- static CirculantGraph(n, adjacency)#
Returns a circulant graph with n nodes.
A circulant graph has the property that the vertex \(i\) is connected with the vertices \(i+j\) and \(i-j\) for each j in
adjacency.INPUT:
n- number of vertices in the graphadjacency- the list of j values
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each circulant graph will be displayed with the first (0) node at the top, with the rest following in a counterclockwise manner.
Filling the position dictionary in advance adds O(n) to the constructor.
See also
sage.graphs.generic_graph.GenericGraph.is_circulant()– checks whether a (di)graph is circulant, and/or returns all possible sets of parameters.
EXAMPLES: Compare plotting using the predefined layout and networkx:
sage: # needs networkx sage: import networkx sage: n = networkx.cycle_graph(23) sage: spring23 = Graph(n) sage: posdict23 = graphs.CirculantGraph(23,2) sage: spring23.show() # long time sage: posdict23.show() # long time
We next view many cycle graphs as a Sage graphics array. First we use the
CirculantGraphconstructor, which fills in the position dictionary:sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.CirculantGraph(i+4, i+1) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
Compare to plotting with the spring-layout algorithm:
sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: spr = networkx.cycle_graph(i+3) ....: k = Graph(spr) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
Passing a 1 into adjacency should give the cycle.
sage: graphs.CirculantGraph(6,1) == graphs.CycleGraph(6) True sage: graphs.CirculantGraph(7,[1,3]).edges(sort=True, labels=false) [(0, 1), (0, 3), (0, 4), (0, 6), (1, 2), (1, 4), (1, 5), (2, 3), (2, 5), (2, 6), (3, 4), (3, 6), (4, 5), (5, 6)]
- static CircularLadderGraph(n)#
Return a circular ladder graph with \(2 * n\) nodes.
A Circular ladder graph is a ladder graph that is connected at the ends, i.e.: a ladder bent around so that top meets bottom. Thus it can be described as two parallel cycle graphs connected at each corresponding node pair.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the circular ladder graph is displayed as an inner and outer cycle pair, with the first \(n\) nodes drawn on the inner circle. The first (0) node is drawn at the top of the inner-circle, moving clockwise after that. The outer circle is drawn with the \((n+1)`th node at the top, then counterclockwise as well. When `n == 2\), we rotate the outer circle by an angle of \(\pi/8\) to ensure that all edges are visible (otherwise the 4 vertices of the graph would be placed on a single line).
EXAMPLES:
Construct and show a circular ladder graph with 26 nodes:
sage: g = graphs.CircularLadderGraph(13) sage: g.show() # long time # needs sage.plot
Create several circular ladder graphs in a Sage graphics array:
sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.CircularLadderGraph(i+3) ....: g.append(k) sage: for i in range(3): # needs sage.plot ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) # needs sage.plot sage: G.show() # long time # needs sage.plot
- static ClawGraph()#
Return a claw graph.
A claw graph is named for its shape. It is actually a complete bipartite graph with
(n1, n2) = (1, 3).PLOTTING: See
CompleteBipartiteGraph().EXAMPLES:
Show a Claw graph:
sage: (graphs.ClawGraph()).show() # long time # needs sage.plot
Inspect a Claw graph:
sage: G = graphs.ClawGraph() sage: G Claw graph: Graph on 4 vertices
- static ClebschGraph()#
Return the Clebsch graph.
See the Wikipedia article Clebsch_graph for more information.
EXAMPLES:
sage: g = graphs.ClebschGraph() sage: g.automorphism_group().cardinality() # needs sage.groups 1920 sage: g.girth() 4 sage: g.chromatic_number() 4 sage: g.diameter() 2 sage: g.show(figsize=[10, 10]) # long time # needs sage.plot
- static CompleteBipartiteGraph(p, q, set_position=True)#
Return a Complete Bipartite Graph on \(p + q\) vertices.
A Complete Bipartite Graph is a graph with its vertices partitioned into two groups, \(V_1 = \{0,...,p-1\}\) and \(V_2 = \{p,...,p+q-1\}\). Each \(u \in V_1\) is connected to every \(v \in V_2\).
INPUT:
p,q– number of vertices in each sideset_position– boolean (defaultTrue); if set toTrue, we assign positions to the vertices so that the set of cardinality \(p\) is on the line \(y=1\) and the set of cardinality \(q\) is on the line \(y=0\).
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each complete bipartite graph will be displayed with the first \(p\) nodes on the top row (at \(y=1\)) from left to right. The remaining \(q\) nodes appear at \(y=0\), also from left to right. The shorter row (partition with fewer nodes) is stretched to the same length as the longer row, unless the shorter row has 1 node; in which case it is centered. The \(x\) values in the plot are in domain \([0, \max(p, q)]\).
In the Complete Bipartite graph, there is a visual difference in using the spring-layout algorithm vs. the position dictionary used in this constructor. The position dictionary flattens the graph and separates the partitioned nodes, making it clear which nodes an edge is connected to. The Complete Bipartite graph plotted with the spring-layout algorithm tends to center the nodes in \(p\) (see
spring_medin examples below), thus overlapping its nodes and edges, making it typically hard to decipher.Filling the position dictionary in advance adds \(O(n)\) to the constructor. Feel free to race the constructors below in the examples section. The much larger difference is the time added by the spring-layout algorithm when plotting. (Also shown in the example below). The spring model is typically described as \(O(n^3)\), as appears to be the case in the NetworkX source code.
EXAMPLES:
Two ways of constructing the complete bipartite graph, using different layout algorithms:
sage: # needs networkx sage: import networkx sage: n = networkx.complete_bipartite_graph(389, 157) # long time sage: spring_big = Graph(n) # long time sage: posdict_big = graphs.CompleteBipartiteGraph(389, 157) # long time
Compare the plotting:
sage: n = networkx.complete_bipartite_graph(11, 17) # needs networkx sage: spring_med = Graph(n) # needs networkx sage: posdict_med = graphs.CompleteBipartiteGraph(11, 17)
Notice here how the spring-layout tends to center the nodes of \(n1\):
sage: spring_med.show() # long time # needs networkx sage: posdict_med.show() # long time # needs sage.plot
View many complete bipartite graphs with a Sage Graphics Array, with this constructor (i.e., the position dictionary filled):
sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.CompleteBipartiteGraph(i+1,4) ....: g.append(k) sage: for i in range(3): # needs sage.plot ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) # needs sage.plot sage: G.show() # long time # needs sage.plot
We compare to plotting with the spring-layout algorithm:
sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: spr = networkx.complete_bipartite_graph(i+1,4) ....: k = Graph(spr) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
sage: graphs.CompleteBipartiteGraph(5,6).complement() complement(Complete bipartite graph of order 5+6): Graph on 11 vertices
- static CompleteGraph(n)#
Return a complete graph on \(n\) nodes.
A Complete Graph is a graph in which all nodes are connected to all other nodes.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each complete graph will be displayed with the first (0) node at the top, with the rest following in a counterclockwise manner.
In the complete graph, there is a big difference visually in using the spring-layout algorithm vs. the position dictionary used in this constructor. The position dictionary flattens the graph, making it clear which nodes an edge is connected to. But the complete graph offers a good example of how the spring-layout works. The edges push outward (everything is connected), causing the graph to appear as a 3-dimensional pointy ball. (See examples below).
EXAMPLES:
We view many Complete graphs with a Sage Graphics Array, first with this constructor (i.e., the position dictionary filled):
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.CompleteGraph(i+3) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
We compare to plotting with the spring-layout algorithm:
sage: # needs networkx sage.plot sage: import networkx sage: g = [] sage: j = [] sage: for i in range(9): ....: spr = networkx.complete_graph(i+3) ....: k = Graph(spr) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
Compare the constructors (results will vary):
sage: # needs networkx sage: import networkx sage: t = cputime() sage: n = networkx.complete_graph(389); spring389 = Graph(n) sage: cputime(t) # random 0.59203700000000126 sage: t = cputime() sage: posdict389 = graphs.CompleteGraph(389) sage: cputime(t) # random 0.6680419999999998
We compare plotting:
sage: # needs networkx sage: import networkx sage: n = networkx.complete_graph(23) sage: spring23 = Graph(n) sage: posdict23 = graphs.CompleteGraph(23) sage: spring23.show() # long time # needs sage.plot sage: posdict23.show() # long time # needs sage.plot
- static CompleteMultipartiteGraph(L)#
Return a complete multipartite graph.
INPUT:
L– a list of integers; the respective sizes of the components
PLOTTING: Produce a layout of the vertices so that vertices in the same vertex set are adjacent and clearly separated from vertices in other vertex sets.
This is done by calculating the vertices of an \(r\)-gon then calculating the slope between adjacent vertices. We then ‘walk’ around the \(r\)-gon placing graph vertices in regular intervals between adjacent vertices of the \(r\)-gon.
Makes a nicely organized graph like in this picture: https://commons.wikimedia.org/wiki/File:Turan_13-4.svg
EXAMPLES:
A complete tripartite graph with sets of sizes \(5, 6, 8\):
sage: g = graphs.CompleteMultipartiteGraph([5, 6, 8]); g Multipartite Graph with set sizes [5, 6, 8]: Graph on 19 vertices
It clearly has a chromatic number of 3:
sage: g.chromatic_number() 3
- static ConwaySmith_for_3S7()#
Return the Conway-Smith graph related to \(3 Sym(7)\).
This is a distance-regular graph with intersection array \([10, 6, 4, 1; 1, 2, 6, 10]\).
EXAMPLES:
sage: G = graphs.ConwaySmith_for_3S7() # needs sage.modules sage.rings.finite_rings sage.rings.number_field sage: G.is_distance_regular(True) # needs sage.modules sage.rings.finite_rings sage.rings.number_field ([10, 6, 4, 1, None], [None, 1, 2, 6, 10])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 399.
- static CorrelationGraph(seqs, alpha, include_anticorrelation)#
Constructs and returns a correlation graph with a node corresponding to each sequence in \(seqs\).
Edges are added between nodes where the corresponding sequences have a correlation coeffecient greater than alpha.
If include_anticorrelation is true, then edges are also added between nodes with correlation coeffecient less than -alpha.
EXAMPLES:
sage: # needs numpy sage: from sage.graphs.generators.basic import CorrelationGraph sage: data = [[1,2,3], [4,5,6], [7,8,9999]] sage: CG1 = CorrelationGraph(data, 0.9, False) sage: CG2 = CorrelationGraph(data, 0.9, True) sage: CG3 = CorrelationGraph(data, 0.1, True) sage: CG1.edges(sort=False) [(0, 0, None), (0, 1, None), (1, 1, None), (2, 2, None)] sage: CG2.edges(sort=False) [(0, 0, None), (0, 1, None), (1, 1, None), (2, 2, None)] sage: CG3.edges(sort=False) [(0, 0, None), (0, 1, None), (0, 2, None), (1, 1, None), (1, 2, None), (2, 2, None)]
- static CossidentePenttilaGraph(q)#
Return the Cossidente-Penttila \(((q^3+1)(q+1)/2,(q^2+1)(q-1)/2,(q-3)/2,(q-1)^2/2)\)-strongly regular graph
For each odd prime power \(q\), one can partition the points of the \(O_6^-(q)\)-generalized quadrangle \(GQ(q,q^2)\) into two parts, so that on any of them the induced subgraph of the point graph of the GQ has parameters as above [CP2005].
Directly following the construction in [CP2005] is not efficient, as one then needs to construct the dual \(GQ(q^2,q)\). Thus we describe here a more efficient approach that we came up with, following a suggestion by T.Penttila. Namely, this partition is invariant under the subgroup \(H=\Omega_3(q^2)<O_6^-(q)\). We build the appropriate \(H\), which leaves the form \(B(X,Y,Z)=XY+Z^2\) invariant, and pick up two orbits of \(H\) on the \(F_q\)-points. One them is \(B\)-isotropic, and we take the representative \((1:0:0)\). The other one corresponds to the points of \(PG(2,q^2)\) that have all the lines on them either missing the conic specified by \(B\), or intersecting the conic in two points. We take \((1:1:e)\) as the representative. It suffices to pick \(e\) so that \(e^2+1\) is not a square in \(F_{q^2}\). Indeed, The conic can be viewed as the union of \(\{(0:1:0)\}\) and \(\{(1:-t^2:t) | t \in F_{q^2}\}\). The coefficients of a generic line on \((1:1:e)\) are \([1:-1-eb:b]\), for \(-1\neq eb\). Thus, to make sure the intersection with the conic is always even, we need that the discriminant of \(1+(1+eb)t^2+tb=0\) never vanishes, and this is if and only if \(e^2+1\) is not a square. Further, we need to adjust \(B\), by multiplying it by appropriately chosen \(\nu\), so that \((1:1:e)\) becomes isotropic under the relative trace norm \(\nu B(X,Y,Z)+(\nu B(X,Y,Z))^q\). The latter is used then to define the graph.
INPUT:
q– an odd prime power.
EXAMPLES:
For \(q=3\) one gets Sims-Gewirtz graph.
sage: G = graphs.CossidentePenttilaGraph(3) # optional - gap_package_grape sage: G.is_strongly_regular(parameters=True) # optional - gap_package_grape (56, 10, 0, 2)
For \(q>3\) one gets new graphs.
sage: G = graphs.CossidentePenttilaGraph(5) # optional - gap_package_grape sage: G.is_strongly_regular(parameters=True) # optional - gap_package_grape (378, 52, 1, 8)
- static CoxeterGraph()#
Return the Coxeter graph.
See the Wikipedia article Coxeter_graph.
EXAMPLES:
sage: g = graphs.CoxeterGraph() sage: g.automorphism_group().cardinality() # needs sage.groups 336 sage: g.girth() 7 sage: g.chromatic_number() 3 sage: g.diameter() 4 sage: g.show(figsize=[10, 10]) # long time # needs sage.plot
- static CubeConnectedCycle(d)#
Return the cube-connected cycle of dimension \(d\).
The cube-connected cycle of order \(d\) is the \(d\)-dimensional hypercube with each of its vertices replaced by a cycle of length \(d\). This graph has order \(d \times 2^d\). The construction is as follows: Construct vertex \((x,y)\) for \(0 \leq x < 2^d\), \(0 \leq y < d\). For each vertex, \((x,y)\), add an edge between it and \((x, (y-1) \mod d))\), \((x,(y+1) \mod d)\), and \((x \oplus 2^y, y)\), where \(\oplus\) is the bitwise xor operator.
For \(d=1\) and \(2\), the cube-connected cycle graph contains self-loops or multiple edges between a pair of vertices, but for all other \(d\), it is simple.
INPUT:
d– The dimension of the desired hypercube as well as the length of the cycle to be placed at each vertex of the \(d\)-dimensional hypercube. \(d\) must be a positive integer.
EXAMPLES:
The order of the graph is \(d \times 2^d\)
sage: d = 3 sage: g = graphs.CubeConnectedCycle(d) sage: len(g) == d*2**d True
The diameter of cube-connected cycles for \(d > 3\) is \(2d + \lfloor \frac{d}{2} \rfloor - 2\)
sage: d = 4 sage: g = graphs.CubeConnectedCycle(d) sage: g.diameter() == 2*d+d//2-2 True
All vertices have degree \(3\) when \(d > 1\)
sage: g = graphs.CubeConnectedCycle(5) sage: all(g.degree(v) == 3 for v in g) True
- static CubeGraph(n, embedding=1)#
Return the \(n\)-cube graph, also called the hypercube in \(n\) dimensions.
The hypercube in \(n\) dimension is build upon the binary strings on \(n\) bits, two of them being adjacent if they differ in exactly one bit. Hence, the distance between two vertices in the hypercube is the Hamming distance.
INPUT:
n– integer; the dimension of the cube graphembedding– integer (default:1); two embeddings of the \(n\)-cube are available:1: the \(n\)-cube is projected inside a regular \(2n\)-gonal polygon by a skew orthogonal projection. See the Wikipedia article Hypercube for more details.2: orthogonal projection of the \(n\)-cube. This orientation shows columns of independent vertices such that the neighbors of a vertex are located in the columns on the left and on the right. The number of vertices in each column represents rows in Pascal’s triangle. See for instance the Wikipedia article 10-cube for more details.3: oblique projection of the \(n\)-cube. Oblique projection involves aligning one face parallel to the viewer and projecting at a specified angle, maintaining equal size for edges parallel to one axis while applying fixed foreshortening to others. This method simplifies the representation of a four-dimensional hypercube onto a two-dimensional plane, offering a geometrically consistent visualization.NoneorO: no embedding is provided
EXAMPLES:
The distance between \(0100110\) and \(1011010\) is \(5\), as expected:
sage: g = graphs.CubeGraph(7) sage: g.distance('0100110','1011010') 5
Plot several \(n\)-cubes in a Sage Graphics Array:
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(6): ....: k = graphs.CubeGraph(i+1) ....: g.append(k) ... sage: for i in range(2): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) ... sage: G = graphics_array(j) sage: G.show(figsize=[6,4]) # long time
Use the plot options to display larger \(n\)-cubes:
sage: g = graphs.CubeGraph(9, embedding=1) sage: g.show(figsize=[12,12],vertex_labels=False, vertex_size=20) # long time, needs sage.plot sage: g = graphs.CubeGraph(9, embedding=2) sage: g.show(figsize=[12,12],vertex_labels=False, vertex_size=20) # long time, needs sage.plot sage: g = graphs.CubeGraph(9, embedding=3) sage: g.show(figsize=[12,12],vertex_labels=False, vertex_size=20) # long time, needs sage.plot
AUTHORS:
Robert Miller
David Coudert
- static CycleGraph(n)#
Return a cycle graph with \(n\) nodes.
A cycle graph is a basic structure which is also typically called an \(n\)-gon.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each cycle graph will be displayed with the first (0) node at the top, with the rest following in a counterclockwise manner.
The cycle graph is a good opportunity to compare efficiency of filling a position dictionary vs. using the spring-layout algorithm for plotting. Because the cycle graph is very symmetric, the resulting plots should be similar (in cases of small \(n\)).
Filling the position dictionary in advance adds \(O(n)\) to the constructor.
EXAMPLES:
Compare plotting using the predefined layout and networkx:
sage: # needs networkx sage.plot sage: import networkx sage: n = networkx.cycle_graph(23) sage: spring23 = Graph(n) sage: posdict23 = graphs.CycleGraph(23) sage: spring23.show() # long time sage: posdict23.show() # long time
We next view many cycle graphs as a Sage graphics array. First we use the
CycleGraphconstructor, which fills in the position dictionary:sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.CycleGraph(i+3) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
Compare to plotting with the spring-layout algorithm:
sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: spr = networkx.cycle_graph(i+3) ....: k = Graph(spr) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static DartGraph()#
Return a dart graph with 5 nodes.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the dart graph is drawn as a dart, with the sharp part on the bottom.
EXAMPLES:
Construct and show a dart graph:
sage: g = graphs.DartGraph() sage: g.show() # long time # needs sage.plot
- static DegreeSequence(deg_sequence)#
Return a graph with the given degree sequence.
This method raises a NetworkX error if the proposed degree sequence cannot be that of a graph.
Graph returned is the one returned by the Havel-Hakimi algorithm, which constructs a simple graph by connecting vertices of highest degree to other vertices of highest degree, resorting the remaining vertices by degree and repeating the process. See Theorem 1.4 in [CL1996].
INPUT:
deg_sequence– list of integers with each entry corresponding to the degree of a different vertex
EXAMPLES:
sage: G = graphs.DegreeSequence([3,3,3,3]) # needs networkx sage: G.edges(sort=True, labels=False) # needs networkx [(0, 1), (0, 2), (0, 3), (1, 2), (1, 3), (2, 3)] sage: G.show() # long time # needs networkx sage.plot
sage: G = graphs.DegreeSequence([3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3]) # needs networkx sage: G.show() # long time # needs networkx sage.plot
sage: G = graphs.DegreeSequence([4,4,4,4,4,4,4,4]) # needs networkx sage: G.show() # long time # needs networkx sage.plot
sage: G = graphs.DegreeSequence([1,2,3,4,3,4,3,2,3,2,1]) # needs networkx sage: G.show() # long time # needs networkx sage.plot
- static DegreeSequenceBipartite(s1, s2)#
Return a bipartite graph whose two sets have the given degree sequences.
Given two different sequences of degrees \(s_1\) and \(s_2\), this functions returns ( if possible ) a bipartite graph on sets \(A\) and \(B\) such that the vertices in \(A\) have \(s_1\) as their degree sequence, while \(s_2\) is the degree sequence of the vertices in \(B\).
INPUT:
s_1– list of integers corresponding to the degree sequence of the first set of verticess_2– list of integers corresponding to the degree sequence of the second set of vertices
ALGORITHM:
This function works through the computation of the matrix given by the Gale-Ryser theorem, which is in this case the adjacency matrix of the bipartite graph.
EXAMPLES:
If we are given as sequences
[2,2,2,2,2]and[5,5]we are given as expected the complete bipartite graph \(K_{2,5}\):sage: g = graphs.DegreeSequenceBipartite([2,2,2,2,2],[5,5]) # needs sage.combinat sage.modules sage: g.is_isomorphic(graphs.CompleteBipartiteGraph(5,2)) # needs sage.combinat sage.modules True
Some sequences being incompatible if, for example, their sums are different, the functions raises a
ValueErrorwhen no graph corresponding to the degree sequences exists:sage: g = graphs.DegreeSequenceBipartite([2,2,2,2,1],[5,5]) # needs sage.combinat sage.modules Traceback (most recent call last): ... ValueError: there exists no bipartite graph corresponding to the given degree sequences
- static DegreeSequenceConfigurationModel(deg_sequence, seed=None)#
Return a random pseudograph with the given degree sequence.
This method raises a NetworkX error if the proposed degree sequence cannot be that of a graph with multiple edges and loops.
One requirement is that the sum of the degrees must be even, since every edge must be incident with two vertices.
INPUT:
deg_sequence– list of integers with each entry corresponding to the expected degree of a different vertexseed– (optional) arandom.Randomseed or a Pythonintfor the random number generator
EXAMPLES:
sage: G = graphs.DegreeSequenceConfigurationModel([1,1]) # needs networkx sage: G.adjacency_matrix() # needs networkx sage.modules [0 1] [1 0]
The output is allowed to contain both loops and multiple edges:
sage: # needs networkx sage: deg_sequence = [3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3] sage: G = graphs.DegreeSequenceConfigurationModel(deg_sequence) sage: G.order(), G.size() (20, 30) sage: G.has_loops() or G.has_multiple_edges() # random True sage: G.show() # long time # needs sage.plot
REFERENCE:
- static DegreeSequenceExpected(deg_sequence, seed=None)#
Return a random graph with expected given degree sequence.
This method raises a NetworkX error if the proposed degree sequence cannot be that of a graph.
One requirement is that the sum of the degrees must be even, since every edge must be incident with two vertices.
INPUT:
deg_sequence– list of integers with each entry corresponding to the expected degree of a different vertexseed– (optional) arandom.Randomseed or a Pythonintfor the random number generator
EXAMPLES:
sage: G = graphs.DegreeSequenceExpected([1,2,3,2,3]); G # needs networkx Looped graph on 5 vertices sage: G.show() # long time # needs networkx sage.plot
REFERENCE:
- static DegreeSequenceTree(deg_sequence)#
Return a tree with the given degree sequence.
This method raises a NetworkX error if the proposed degree sequence cannot be that of a tree.
Since every tree has one more vertex than edge, the degree sequence must satisfy
len(deg_sequence) - sum(deg_sequence)/2 == 1.INPUT:
deg_sequence– list of integers with each entry corresponding to the expected degree of a different vertex
EXAMPLES:
sage: G = graphs.DegreeSequenceTree([3,1,3,3,1,1,1,2,1]); G # needs networkx Graph on 9 vertices sage: G.show() # long time # needs networkx sage.plot
- static DejterGraph()#
Return the Dejter graph.
The Dejter graph is obtained from the binary 7-cube by deleting a copy of the Hamming code of length 7. It is 6-regular, with 112 vertices and 336 edges. For more information, see the Wikipedia article Dejter_graph.
EXAMPLES:
sage: g = graphs.DejterGraph(); g # needs sage.rings.finite_rings Dejter Graph: Graph on 112 vertices sage: g.is_regular(k=6) # needs sage.rings.finite_rings True sage: g.girth() # needs sage.rings.finite_rings 4
- static DesarguesGraph()#
Return the Desargues graph.
PLOTTING: The layout chosen is the same as on the cover of [Har1994].
EXAMPLES:
sage: D = graphs.DesarguesGraph() sage: L = graphs.LCFGraph(20,[5,-5,9,-9],5) # needs networkx sage: D.is_isomorphic(L) # needs networkx True sage: D.show() # long time # needs sage.plot
- static DiamondGraph()#
Return a diamond graph with 4 nodes.
A diamond graph is a square with one pair of diagonal nodes connected.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the diamond graph is drawn as a diamond, with the first node on top, second on the left, third on the right, and fourth on the bottom; with the second and third node connected.
EXAMPLES:
Construct and show a diamond graph:
sage: g = graphs.DiamondGraph() sage: g.show() # long time # needs sage.plot
- static DipoleGraph(n)#
Returns a dipole graph with n edges.
A dipole graph is a multigraph consisting of 2 vertices connected with n parallel edges.
EXAMPLES:
Construct and show a dipole graph with 13 edges:
sage: g = graphs.DipoleGraph(13); g Dipole graph: Multi-graph on 2 vertices sage: g.show() # long time # needs sage.plot
- static DodecahedralGraph()#
Return a Dodecahedral graph (with 20 nodes)
The dodecahedral graph is cubic symmetric, so the spring-layout algorithm will be very effective for display. It is dual to the icosahedral graph.
PLOTTING: The Dodecahedral graph should be viewed in 3 dimensions. We choose to use a planar embedding of the graph. We hope to add rotatable, 3-dimensional viewing in the future. In such a case, a argument will be added to select the desired layout.
EXAMPLES:
Construct and show a Dodecahedral graph:
sage: g = graphs.DodecahedralGraph() sage: g.show() # long time # needs sage.plot
Create several dodecahedral graphs in a Sage graphics array They will be drawn differently due to the use of the spring-layout algorithm:
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.DodecahedralGraph() ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static DorogovtsevGoltsevMendesGraph(n)#
Construct the n-th generation of the Dorogovtsev-Goltsev-Mendes graph.
EXAMPLES:
sage: G = graphs.DorogovtsevGoltsevMendesGraph(8) # needs networkx sage: G.size() # needs networkx 6561
REFERENCE:
[1] Dorogovtsev, S. N., Goltsev, A. V., and Mendes, J. F. F., Pseudofractal scale-free web, Phys. Rev. E 066122 (2002).
- static DoubleGeneralizedPetersenGraph(n, k)#
Return a double generalized Petersen graph with \(4n\) nodes.
The double generalized Petersen graphs is a family of graphs proposed in [ZF2012] as a variant of generalized Petersen graphs. The variables \(n\), \(k\) are integers such that \(n > 2\) and \(0 < k \leq \lfloor (n-1) / 2 \rfloor\).
INPUT:
n– the number of nodes is \(4 * n\)k– integer such that \(0 < k \leq \lfloor (n-1) / 2 \rfloor\) determining how vertices on second and third inner rims are connected
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the double generalized Petersen graphs are displayed as 4 cocentric cycles, with the first n nodes drawn on the outer circle. The first (0) node is drawn at the top of the outer-circle, moving counterclockwise after that. The second circle is drawn with the (n)th node at the top, then counterclockwise as well. The tird cycle is drawn with the (2n)th node at the top, then counterclockwise. And the fourth cycle is drawn with the (3n)th node at the top, then again counterclockwise.
EXAMPLES:
When \(n\) is even the resulting graph will be isomorphic to a double generalized Petersen graph with \(k' = n / 2 - k\):
sage: g = graphs.DoubleGeneralizedPetersenGraph(10, 2) sage: g2 = graphs.DoubleGeneralizedPetersenGraph(10, 3) sage: g.is_isomorphic(g2) True
- static DoubleGrassmannGraph(q, e)#
Return the bipartite double of the distance-\(e\) graph of the Grassmann graph \(J_q(n,e)\).
This graph can also be descirbed as follows: Let \(V\) be the vector space of dimension \(n\) over \(GF(q)\). The vertex set is the set of \(e+1\) or \(e\) subspaces of \(V\). Two vertices are adjacent if one subspace is contained in the other.
This graph is distance-transitive.
INPUT:
q– a prime powere– integer
EXAMPLES:
sage: G = graphs.DoubleGrassmannGraph(2,1) # needs sage.modules sage: G.diameter() # needs sage.modules 3 sage: G.is_distance_regular(True) # needs sage.modules ([3, 2, 2, None], [None, 1, 1, 3])
REFERENCES:
See [BCN1989] pp. 272, 273 or [VDKT2016] p. 25.
- static DoubleOddGraph(n)#
Return the double odd graph on \(2n+1\) points.
The graph is obtained using the subsets of size \(n\) and \(n+1\) of \({1, 2, ..., 2n+1}\) as vertices. Two vertices are adjacent if one is included in the other.
The graph is distance-transitive.
INPUT:
n– integer; must be greater than 0
EXAMPLES:
sage: G = graphs.DoubleOddGraph(5) sage: G.is_distance_regular(True) ([6, 5, 5, 4, 4, 3, 3, 2, 2, 1, 1, None], [None, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6]) sage: G = graphs.DoubleOddGraph(3) sage: G.diameter() 7 sage: G.is_distance_regular(True) ([4, 3, 3, 2, 2, 1, 1, None], [None, 1, 1, 2, 2, 3, 3, 4])
REFERENCES:
See [BCN1989] pp. 259-261 or [VDKT2016] p. 25.
- static DoubleStarSnark()#
Return the double star snark.
The double star snark is a 3-regular graph on 30 vertices. See the Wikipedia article Double-star_snark.
EXAMPLES:
sage: g = graphs.DoubleStarSnark() sage: g.order() 30 sage: g.size() 45 sage: g.chromatic_number() 3 sage: g.is_hamiltonian() # needs sage.numerical.mip False sage: g.automorphism_group().cardinality() # needs sage.groups 80 sage: g.show() # needs sage.plot
- static DoublyTruncatedWittGraph()#
Return the doubly truncated Witt graph.
This builds the truncated Witt graph, then removes all vertices whose codeword start with a 1.
The graph is distance-regular with intersection array \([7,6,4,4;1,1,1,6]\).
EXAMPLES:
sage: G = graphs.DoublyTruncatedWittGraph() # needs sage.libs.pari sage.modules sage: G.is_distance_regular(True) # needs sage.libs.pari sage.modules ([7, 6, 4, 4, None], [None, 1, 1, 1, 6])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 368.
- static DurerGraph()#
Return the Dürer graph.
For more information, see the Wikipedia article D%C3%BCrer_graph.
EXAMPLES:
The Dürer graph is named after Albrecht Dürer. It is a planar graph with 12 vertices and 18 edges:
sage: G = graphs.DurerGraph(); G Durer graph: Graph on 12 vertices sage: G.is_planar() True sage: G.order() 12 sage: G.size() 18
The Dürer graph has chromatic number 3, diameter 4, and girth 3:
sage: G.chromatic_number() 3 sage: G.diameter() 4 sage: G.girth() 3
Its automorphism group is isomorphic to \(D_6\):
sage: ag = G.automorphism_group() # needs sage.groups sage: ag.is_isomorphic(DihedralGroup(6)) # needs sage.groups True
- static DyckGraph()#
Return the Dyck graph.
For more information, see the MathWorld article on the Dyck graph or the Wikipedia article Dyck_graph.
EXAMPLES:
The Dyck graph was defined by Walther von Dyck in 1881. It has \(32\) vertices and \(48\) edges, and is a cubic graph (regular of degree \(3\)):
sage: G = graphs.DyckGraph(); G Dyck graph: Graph on 32 vertices sage: G.order() 32 sage: G.size() 48 sage: G.is_regular() True sage: G.is_regular(3) True
It is non-planar and Hamiltonian, as well as bipartite (making it a bicubic graph):
sage: G.is_planar() False sage: G.is_hamiltonian() # needs sage.numerical.mip True sage: G.is_bipartite() True
It has radius \(5\), diameter \(5\), and girth \(6\):
sage: G.radius() 5 sage: G.diameter() 5 sage: G.girth() 6
Its chromatic number is \(2\) and its automorphism group is of order \(192\):
sage: G.chromatic_number() 2 sage: G.automorphism_group().cardinality() # needs sage.groups 192
It is a non-integral graph as it has irrational eigenvalues:
sage: G.characteristic_polynomial().factor() # needs sage.libs.pari sage.modules (x - 3) * (x + 3) * (x - 1)^9 * (x + 1)^9 * (x^2 - 5)^6
It is a toroidal graph, and its embedding on a torus is dual to an embedding of the Shrikhande graph (
ShrikhandeGraph).
- static EgawaGraph(p, s)#
Return the Egawa graph with parameters \(p\), \(s\).
Egawa graphs are a peculiar family of graphs devised by Yoshimi Egawa in [Ega1981] . The Shrikhande graph is a special case of this family of graphs, with parameters \((1,0)\). All the graphs in this family are not recognizable by 1-WL (Weisfeiler Lehamn algorithm of the first order) and 2-WL, that is their orbits are not correctly returned by k-WL for k lower than 3.
Furthermore, all the graphs in this family are distance-regular, but they are not distance-transitive if \(p \neq 0\).
The Egawa graph with parameters \((0, s)\) is isomorphic to the Hamming graph with parameters \((s, 4)\), when the underlying set of the Hamming graph is \([0,1,2,3]\)
INPUT:
p– power to which the graph named \(Y\) in the referenceprovided above will be raised
s– power to which the graph named \(X\) in the referenceprovided above will be raised
OUTPUT:
G– The Egawa graph with parameters (p,s)
EXAMPLES:
Every Egawa graph is distance regular.
sage: g = graphs.EgawaGraph(1, 2) sage: g.is_distance_regular() True
An Egawa graph with parameters (0,s) is isomorphic to the Hamming graph with parameters (s, 4).
sage: g = graphs.EgawaGraph(0, 4) sage: g.is_isomorphic(graphs.HammingGraph(4,4)) True
- static EllinghamHorton54Graph()#
Return the Ellingham-Horton 54-graph.
For more information, see the Wikipedia article Ellingham-Horton_graph.
EXAMPLES:
This graph is 3-regular:
sage: g = graphs.EllinghamHorton54Graph() sage: g.is_regular(k=3) True
It is 3-connected and bipartite:
sage: g.vertex_connectivity() # not tested - too long 3 sage: g.is_bipartite() True
It is not Hamiltonian:
sage: g.is_hamiltonian() # not tested # needs sage.numerical.mip False
… and it has a nice drawing
sage: g.show(figsize=[10, 10]) # not tested - too long
- static EllinghamHorton78Graph()#
Return the Ellingham-Horton 78-graph.
For more information, see the Wikipedia article Ellingham%E2%80%93Horton_graph
EXAMPLES:
This graph is 3-regular:
sage: g = graphs.EllinghamHorton78Graph() sage: g.is_regular(k=3) True
It is 3-connected and bipartite:
sage: g.vertex_connectivity() # not tested (too long) 3 sage: g.is_bipartite() True
It is not Hamiltonian:
sage: g.is_hamiltonian() # not tested # needs sage.numerical.mip False
… and it has a nice drawing
sage: g.show(figsize=[10,10]) # not tested (too long)
- static EmptyGraph()#
Return an empty graph (0 nodes and 0 edges).
This is useful for constructing graphs by adding edges and vertices individually or in a loop.
PLOTTING: When plotting, this graph will use the default spring-layout algorithm, unless a position dictionary is specified.
EXAMPLES:
Add one vertex to an empty graph and then show:
sage: empty1 = graphs.EmptyGraph() sage: empty1.add_vertex() 0 sage: empty1.show() # long time # needs sage.plot
Use for loops to build a graph from an empty graph:
sage: empty2 = graphs.EmptyGraph() sage: for i in range(5): ....: empty2.add_vertex() # add 5 nodes, labeled 0-4 0 1 2 3 4 sage: for i in range(3): ....: empty2.add_edge(i,i+1) # add edges {[0:1],[1:2],[2:3]} sage: for i in range(1, 4): ....: empty2.add_edge(4,i) # add edges {[1:4],[2:4],[3:4]} sage: empty2.show() # long time # needs sage.plot
- static ErreraGraph()#
Return the Errera graph.
For more information, see the Wikipedia article Errera_graph.
EXAMPLES:
The Errera graph is named after Alfred Errera. It is a planar graph on 17 vertices and having 45 edges:
sage: G = graphs.ErreraGraph(); G Errera graph: Graph on 17 vertices sage: G.is_planar() True sage: G.order() 17 sage: G.size() 45
The Errera graph is Hamiltonian with radius 3, diameter 4, girth 3, and chromatic number 4:
sage: G.is_hamiltonian() # needs sage.numerical.mip True sage: G.radius() 3 sage: G.diameter() 4 sage: G.girth() 3 sage: G.chromatic_number() 4
Each vertex degree is either 5 or 6. That is, if \(f\) counts the number of vertices of degree 5 and \(s\) counts the number of vertices of degree 6, then \(f + s\) is equal to the order of the Errera graph:
sage: D = G.degree_sequence() sage: D.count(5) + D.count(6) == G.order() True
The automorphism group of the Errera graph is isomorphic to the dihedral group of order 20:
sage: ag = G.automorphism_group() # needs sage.groups sage: ag.is_isomorphic(DihedralGroup(10)) # needs sage.groups True
- static EuropeMap(continental=False, year=2018)#
Return European states as a graph of common border.
“European state” here is defined as an independent state having the capital city in Europe. The graph has an edge between those countries that have common land border.
INPUT:
continental– boolean (default:False); whether to only return states in the continental Europe or all European statesyear– integer (default:2018); reserved for future use
EXAMPLES:
sage: Europe = graphs.EuropeMap(); Europe Europe Map: Graph on 44 vertices sage: Europe.neighbors('Ireland') ['United Kingdom'] sage: cont_Europe = graphs.EuropeMap(continental=True) sage: cont_Europe.order() 40 sage: 'Iceland' in cont_Europe False
- static F26AGraph()#
Return the F26A graph.
The F26A graph is a symmetric bipartite cubic graph with 26 vertices and 39 edges. For more information, see the Wikipedia article F26A_graph.
EXAMPLES:
sage: # needs networkx sage: g = graphs.F26AGraph(); g F26A Graph: Graph on 26 vertices sage: g.order(), g.size() (26, 39) sage: g.automorphism_group().cardinality() 78 sage: g.girth() 6 sage: g.is_bipartite() True sage: g.characteristic_polynomial().factor() (x - 3) * (x + 3) * (x^4 - 5*x^2 + 3)^6
- static FibonacciTree(n)#
Return the graph of the Fibonacci Tree \(F_{i}\) of order \(n\).
The Fibonacci tree \(F_{i}\) is recursively defined as the tree with a root vertex and two attached child trees \(F_{i-1}\) and \(F_{i-2}\), where \(F_{1}\) is just one vertex and \(F_{0}\) is empty.
INPUT:
n- the recursion depth of the Fibonacci Tree
EXAMPLES:
sage: g = graphs.FibonacciTree(3) # needs sage.libs.pari sage: g.is_tree() # needs sage.libs.pari True
sage: l1 = [ len(graphs.FibonacciTree(_)) + 1 for _ in range(6) ] # needs sage.libs.pari sage: l2 = list(fibonacci_sequence(2,8)) # needs sage.libs.pari sage: l1 == l2 # needs sage.libs.pari True
AUTHORS:
Harald Schilly and Yann Laigle-Chapuy (2010-03-25)
- static FlowerSnark()#
Return a Flower Snark.
A flower snark has 20 vertices. It is part of the class of biconnected cubic graphs with edge chromatic number = 4, known as snarks. (i.e.: the Petersen graph). All snarks are not Hamiltonian, non-planar and have Petersen graph graph minors. See the Wikipedia article Flower_snark.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the nodes are drawn 0-14 on the outer circle, and 15-19 in an inner pentagon.
EXAMPLES: Inspect a flower snark:
sage: F = graphs.FlowerSnark() sage: F Flower Snark: Graph on 20 vertices sage: F.graph6_string() 'ShCGHC@?GGg@?@?Gp?K??C?CA?G?_G?Cc'
Now show it:
sage: F.show() # long time # needs sage.plot
- static FoldedCubeGraph(n)#
Returns the folded cube graph of order \(2^{n-1}\).
The folded cube graph on \(2^{n-1}\) vertices can be obtained from a cube graph on \(2^n\) vertices by merging together opposed vertices. Alternatively, it can be obtained from a cube graph on \(2^{n-1}\) vertices by adding an edge between opposed vertices. This second construction is the one produced by this method.
See the Wikipedia article Folded_cube_graph for more information.
EXAMPLES:
The folded cube graph of order five is the Clebsch graph:
sage: fc = graphs.FoldedCubeGraph(5) sage: clebsch = graphs.ClebschGraph() sage: fc.is_isomorphic(clebsch) True
- static FolkmanGraph()#
Return the Folkman graph.
See the Wikipedia article Folkman_graph.
EXAMPLES:
sage: # needs networkx sage: g = graphs.FolkmanGraph() sage: g.order() 20 sage: g.size() 40 sage: g.diameter() 4 sage: g.girth() 4 sage: g.charpoly().factor() (x - 4) * (x + 4) * x^10 * (x^2 - 6)^4 sage: g.chromatic_number() 2 sage: g.is_eulerian() True sage: g.is_hamiltonian() # needs sage.numerical_mip True sage: g.is_vertex_transitive() False sage: g.is_bipartite() True
- static ForkGraph()#
Return a fork graph with 5 nodes.
A fork graph, sometimes also called chair graph, is 5 vertex tree.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the fork graph is drawn as a fork, with the sharp part on the bottom.
EXAMPLES:
Construct and show a fork graph:
sage: g = graphs.ForkGraph() sage: g.show() # long time # needs sage.plot
- static FosterGraph()#
Return the Foster graph.
See the Wikipedia article Foster_graph.
EXAMPLES:
sage: # needs networkx sage: g = graphs.FosterGraph() sage: g.order() 90 sage: g.size() 135 sage: g.diameter() 8 sage: g.girth() 10 sage: g.automorphism_group().cardinality() 4320 sage: g.is_hamiltonian() # needs sage.numerical_mip True
- static FosterGraph3S6()#
Return the Foster graph for \(3.Sym(6)\).
This graph is distance-regular with intersection array \([6, 4, 2, 1; 1, 1, 4, 6]\).
The graph is also distance transitive.
EXAMPLES:
sage: G = graphs.FosterGraph3S6() # needs sage.libs.gap sage: G.is_distance_regular(True) # needs sage.libs.gap ([6, 4, 2, 1, None], [None, 1, 1, 4, 6])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 397.
- static FranklinGraph()#
Return the Franklin graph.
For more information, see the Wikipedia article Franklin_graph.
EXAMPLES:
The Franklin graph is named after Philip Franklin. It is a 3-regular graph on 12 vertices and having 18 edges:
sage: G = graphs.FranklinGraph(); G Franklin graph: Graph on 12 vertices sage: G.is_regular(3) True sage: G.order() 12 sage: G.size() 18
The Franklin graph is a Hamiltonian, bipartite graph with radius 3, diameter 3, and girth 4:
sage: G.is_hamiltonian() # needs sage.numerical_mip True sage: G.is_bipartite() True sage: G.radius() 3 sage: G.diameter() 3 sage: G.girth() 4
It is a perfect, triangle-free graph having chromatic number 2:
sage: G.is_perfect() True sage: G.is_triangle_free() True sage: G.chromatic_number() 2
- static FriendshipGraph(n)#
Return the friendship graph \(F_n\).
The friendship graph is also known as the Dutch windmill graph. Let \(C_3\) be the cycle graph on 3 vertices. Then \(F_n\) is constructed by joining \(n \geq 1\) copies of \(C_3\) at a common vertex. If \(n = 1\), then \(F_1\) is isomorphic to \(C_3\) (the triangle graph). If \(n = 2\), then \(F_2\) is the butterfly graph, otherwise known as the bowtie graph. For more information, see the Wikipedia article Friendship_graph.
INPUT:
n– positive integer; the number of copies of \(C_3\) to use in constructing \(F_n\).
OUTPUT:
The friendship graph \(F_n\) obtained from \(n\) copies of the cycle graph \(C_3\).
See also
EXAMPLES:
The first few friendship graphs.
sage: # needs sage.plot sage: A = []; B = [] sage: for i in range(9): ....: g = graphs.FriendshipGraph(i + 1) ....: A.append(g) sage: for i in range(3): ....: n = [] ....: for j in range(3): ....: n.append(A[3*i + j].plot(vertex_size=20, vertex_labels=False)) ....: B.append(n) sage: G = graphics_array(B) sage: G.show() # long time
For \(n = 1\), the friendship graph \(F_1\) is isomorphic to the cycle graph \(C_3\), whose visual representation is a triangle.
sage: G = graphs.FriendshipGraph(1); G Friendship graph: Graph on 3 vertices sage: G.show() # long time # needs sage.plot sage: G.is_isomorphic(graphs.CycleGraph(3)) True
For \(n = 2\), the friendship graph \(F_2\) is isomorphic to the butterfly graph, otherwise known as the bowtie graph.
sage: G = graphs.FriendshipGraph(2); G Friendship graph: Graph on 5 vertices sage: G.is_isomorphic(graphs.ButterflyGraph()) True
If \(n \geq 2\), then the friendship graph \(F_n\) has \(2n + 1\) vertices and \(3n\) edges. It has radius 1, diameter 2, girth 3, and chromatic number 3. Furthermore, \(F_n\) is planar and Eulerian.
sage: n = randint(2, 10^3) sage: G = graphs.FriendshipGraph(n) sage: G.order() == 2*n + 1 True sage: G.size() == 3*n True sage: G.radius() 1 sage: G.diameter() 2 sage: G.girth() 3 sage: G.chromatic_number() 3 sage: G.is_planar() True sage: G.is_eulerian() True
- static FruchtGraph()#
Return a Frucht Graph.
A Frucht graph has 12 nodes and 18 edges. It is the smallest cubic identity graph. It is planar and Hamiltonian. See the Wikipedia article Frucht_graph.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the first seven nodes are on the outer circle, with the next four on an inner circle and the last in the center.
EXAMPLES:
sage: FRUCHT = graphs.FruchtGraph() sage: FRUCHT Frucht graph: Graph on 12 vertices sage: FRUCHT.graph6_string() 'KhCKM?_EGK?L' sage: (graphs.FruchtGraph()).show() # long time # needs networkx
- static FurerGadget(k, prefix=None)#
Return a Furer gadget of order
kand their coloring.Construct the Furer gadget described in [CFI1992], a graph composed by a middle layer of \(2^(k-1)\) nodes and two sets of nodes \((a_0, ... , a_{k-1})\) and \((b_0, ... , b_{k-1})\). Each node in the middle is connected to either \(a_i\) or \(b_i\), for each i in [0,k[. To read about the complete construction, see [CFI1992]. The returned coloring colors the middle section with one color, and then each pair \((a_i, b_i)\) with another color. Since this method is mainly used to create Furer gadgets for the Cai-Furer-Immerman construction, returning gadgets that don’t always have the same vertex labels is important, that’s why there is a parameter to manually set a prefix to be appended to each vertex label.
INPUT:
k– The order of the returned Furer gadget, greater than 0.prefix– Prefix of to be appended to each vertex label,so as to individualise the returned Furer gadget. Must be comparable for equality and hashable.
OUTPUT:
G– The Furer gadget of orderkcoloring– A list of list of vertices, representing thepartition induced by the coloring of
G’s vertices
EXAMPLES:
Furer gadget of order 3, without any prefix.
sage: G, p = graphs.FurerGadget(3) sage: sorted(G, key=str) [(), (0, 'a'), (0, 'b'), (0, 1), (0, 2), (1, 'a'), (1, 'b'), (1, 2), (2, 'a'), (2, 'b')] sage: sorted(G.edge_iterator(), key=str) [((), (0, 'b'), None), ((), (1, 'b'), None), ((), (2, 'b'), None), ((0, 'b'), (1, 2), None), ((0, 1), (0, 'a'), None), ((0, 1), (1, 'a'), None), ((0, 1), (2, 'b'), None), ((0, 2), (0, 'a'), None), ((0, 2), (1, 'b'), None), ((0, 2), (2, 'a'), None), ((1, 2), (1, 'a'), None), ((1, 2), (2, 'a'), None)]
Furer gadget of order 3, with a prefix.
sage: G, p = graphs.FurerGadget(3, 'Prefix') sage: sorted(G, key=str) [('Prefix', ()), ('Prefix', (0, 'a')), ('Prefix', (0, 'b')), ('Prefix', (0, 1)), ('Prefix', (0, 2)), ('Prefix', (1, 'a')), ('Prefix', (1, 'b')), ('Prefix', (1, 2)), ('Prefix', (2, 'a')), ('Prefix', (2, 'b'))] sage: sorted(G.edge_iterator(), key=str) [(('Prefix', ()), ('Prefix', (0, 'b')), None), (('Prefix', ()), ('Prefix', (1, 'b')), None), (('Prefix', ()), ('Prefix', (2, 'b')), None), (('Prefix', (0, 'b')), ('Prefix', (1, 2)), None), (('Prefix', (0, 1)), ('Prefix', (0, 'a')), None), (('Prefix', (0, 1)), ('Prefix', (1, 'a')), None), (('Prefix', (0, 1)), ('Prefix', (2, 'b')), None), (('Prefix', (0, 2)), ('Prefix', (0, 'a')), None), (('Prefix', (0, 2)), ('Prefix', (1, 'b')), None), (('Prefix', (0, 2)), ('Prefix', (2, 'a')), None), (('Prefix', (1, 2)), ('Prefix', (1, 'a')), None), (('Prefix', (1, 2)), ('Prefix', (2, 'a')), None)]
- static FuzzyBallGraph(partition, q)#
Construct a Fuzzy Ball graph with the integer partition
partitionandqextra vertices.Let \(q\) be an integer and let \(m_1,m_2,...,m_k\) be a set of positive integers. Let \(n=q+m_1+...+m_k\). The Fuzzy Ball graph with partition \(m_1,m_2,...,m_k\) and \(q\) extra vertices is the graph constructed from the graph \(G=K_n\) by attaching, for each \(i=1,2,...,k\), a new vertex \(a_i\) to \(m_i\) distinct vertices of \(G\).
For given positive integers \(k\) and \(m\) and nonnegative integer \(q\), the set of graphs
FuzzyBallGraph(p, q)for all partitions \(p\) of \(m\) with \(k\) parts are cospectral with respect to the normalized Laplacian.EXAMPLES:
sage: F = graphs.FuzzyBallGraph([3,1],2) sage: F.adjacency_matrix(vertices=list(F)) # needs sage.modules [0 0 1 1 1 0 0 0] [0 0 0 0 0 1 0 0] [1 0 0 1 1 1 1 1] [1 0 1 0 1 1 1 1] [1 0 1 1 0 1 1 1] [0 1 1 1 1 0 1 1] [0 0 1 1 1 1 0 1] [0 0 1 1 1 1 1 0]
Pick positive integers \(m\) and \(k\) and a nonnegative integer \(q\). All the FuzzyBallGraphs constructed from partitions of \(m\) with \(k\) parts should be cospectral with respect to the normalized Laplacian:
sage: m = 4; q = 2; k = 2 sage: g_list = [graphs.FuzzyBallGraph(p,q) # needs sage.combinat sage.modules ....: for p in Partitions(m, length=k)] sage: set(g.laplacian_matrix(normalized=True, # long time (7s on sage.math, 2011), needs sage.combinat sage.modules ....: vertices=list(g)).charpoly() ....: for g in g_list) {x^8 - 8*x^7 + 4079/150*x^6 - 68689/1350*x^5 + 610783/10800*x^4 - 120877/3240*x^3 + 1351/100*x^2 - 931/450*x}
- static GemGraph()#
Return a gem graph with 5 nodes.
A gem graph is a fan graph (4,1).
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the gem graph is drawn as a gem, with the sharp part on the bottom.
EXAMPLES:
Construct and show a gem graph:
sage: g = graphs.GemGraph() sage: g.show() # long time # needs sage.plot
- static GeneralisedDodecagonGraph(s, t)#
Return the point-graph of a generalised dodecagon of order \((s,t)\).
INPUT:
s, t– integers; order of the generalised dodecagon
EXAMPLES:
sage: # optional - gap_package_atlasrep internet sage: G = graphs.GeneralisedDodecagonGraph(1, 5) sage: G.is_distance_regular(True) ([6, 5, 5, 5, 5, 5, None], [None, 1, 1, 1, 1, 1, 6]) sage: H = graphs.GeneralisedDodecagonGraph(5, 1) sage: H.order() 23436 sage: H.is_distance_regular(True) # not tested (6 min) ([10, 5, 5, 5, 5, 5, None], [None, 1, 1, 1, 1, 1, 2])
Note
This function indirectly uses the GAP’s AtlasRep package. Thus you may need an internet connection and the optional Sage’s package
gap_packages.REFERENCES:
See [BCN1989] pp. 200-205 for a discussion of distance-regular graphs from generalised polygons.
- static GeneralisedHexagonGraph(s, t)#
Return the point-graph of a generalised hexagon of order \((s,t)\).
INPUT:
s, t– integers; order of the generalised hexagon
EXAMPLES:
sage: # needs sage.libs.gap sage: G = graphs.GeneralisedHexagonGraph(5, 5) # optional - gap_package_atlasrep internet sage: G.is_distance_regular(True) # optional - gap_package_atlasrep internet ([30, 25, 25, None], [None, 1, 1, 6]) sage: G = graphs.GeneralisedHexagonGraph(7, 1) sage: G.is_distance_regular(True) ([14, 7, 7, None], [None, 1, 1, 2]) sage: graphs.GeneralisedHexagonGraph(1, 1) Cycle graph: Graph on 6 vertices
Note
This function uses the GAP’s AtlasRep package to build GHs of order \((q, q)\), \((q, q^3)\) or \((q^3, q)\). For those graphs you need an internet connection and Sage’s optional package
gap_packages.REFERENCES:
See [BCN1989] pp. 200-205 for a discussion of distance-regular graphs from generalised polygons.
- static GeneralisedOctagonGraph(s, t)#
Return the point-graph of a generalised octagon of order \((s,t)\).
INPUT:
s, t– integers; order of the generalised octagon
EXAMPLES:
sage: # needs sage.libs.gap sage: G = graphs.GeneralisedOctagonGraph(1, 4) sage: G.is_distance_regular(True) ([5, 4, 4, 4, None], [None, 1, 1, 1, 5]) sage: G = graphs.GeneralisedOctagonGraph(2, 4) # optional - gap_package_atlasrep internet sage: G.is_distance_regular(True) # optional - gap_package_atlasrep internet ([10, 8, 8, 8, None], [None, 1, 1, 1, 5]) sage: G = graphs.GeneralisedOctagonGraph(5, 1) sage: G.is_distance_regular(True) ([10, 5, 5, 5, None], [None, 1, 1, 1, 2])
Note
This function uses the GAP’s AtlasRep package to build the graphs of order \((2, 4)\) or \((4, 2)\). For those graphs you need an internet connection and Sage’s optional package
gap_packages.REFERENCES:
See [BCN1989] pp. 200-205 for a discussion of distance-regular graphs from generalised polygons.
- static GeneralizedPetersenGraph(n, k)#
Returns a generalized Petersen graph with \(2n\) nodes. The variables \(n\), \(k\) are integers such that \(n>2\) and \(0<k\leq\lfloor(n-1)\)/\(2\rfloor\)
For \(k=1\) the result is a graph isomorphic to the circular ladder graph with the same \(n\). The regular Petersen Graph has \(n=5\) and \(k=2\). Other named graphs that can be described using this notation include the Desargues graph and the Möbius-Kantor graph.
INPUT:
n- the number of nodes is \(2*n\).k- integer \(0<k\leq\lfloor(n-1)\)/\(2\rfloor\). Decides how inner vertices are connected.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the generalized Petersen graphs are displayed as an inner and outer cycle pair, with the first n nodes drawn on the outer circle. The first (0) node is drawn at the top of the outer-circle, moving counterclockwise after that. The inner circle is drawn with the (n)th node at the top, then counterclockwise as well.
EXAMPLES: For \(k=1\) the resulting graph will be isomorphic to a circular ladder graph.
sage: g = graphs.GeneralizedPetersenGraph(13,1) sage: g2 = graphs.CircularLadderGraph(13) sage: g.is_isomorphic(g2) True
The Desargues graph:
sage: g = graphs.GeneralizedPetersenGraph(10,3) sage: g.girth() 6 sage: g.is_bipartite() True
AUTHORS:
Anders Jonsson (2009-10-15)
- static GeneralizedSierpinskiGraph(G, k, stretch=None)#
Return the generalized Sierpinski graph of \(G\) of dimension \(k\).
Generalized Sierpinski graphs have been introduced in [GKP2011] to generalize the notion of Sierpinski graphs [KM1997].
Given a graph \(G = (V, E)\) of order \(n\) and a parameter \(k\), the generalized Sierpinski graph of \(G\) of dimension \(k\), denoted by \(S(G, k)\), can be constructed recursively from \(G\) as follows. \(S(G, 1)\) is isomorphic to \(G\). To construct \(S(G, k)\) for \(k > 1\), copy \(n\) times \(S(G, k - 1)\), once per vertex \(u \in V\), and add \(u\) at the beginning of the labels of each vertex in the copy of \(S(G, k - 1)\) corresponding to vertex \(u\). Then for any edge \(\{u, v\} \in E\), add an edge between vertex \((u, v, \ldots, v)\) and vertex \((v, u, \ldots, u)\).
INPUT:
G– a sage Graphk– integer; the dimensionstretch– integer (default:None); stretching factor used to determine the positions of the vertices of the output graph. By default (None), this value is set to twice the maximum Euclidean distance between the vertices of \(G\). This parameter is used only when the vertices of \(G\) have positions.
See also
EXAMPLES:
The generalized Sierpinski graph of dimension 1 of any graph \(G\) is isomorphic to \(G\):
sage: G = graphs.RandomGNP(10, .5) sage: S = graphs.GeneralizedSierpinskiGraph(G, 1) sage: S.is_isomorphic(G) True
When \(G\) is a clique of order 3, the generalized Sierpinski graphs of \(G\) are isomorphic to Hanoi Tower graphs:
sage: k = randint(1, 5) sage: S = graphs.GeneralizedSierpinskiGraph(graphs.CompleteGraph(3), k) # needs sage.modules sage: H = graphs.HanoiTowerGraph(3, k) sage: S.is_isomorphic(H) # needs sage.modules True
The generalized Sierpinski graph of dimension \(k\) of any graph \(G\) with \(n\) vertices and \(m\) edges has \(n^k\) vertices and \(m\sum_{i=0}^{k-1}n^i\) edges:
sage: # needs sage.modules sage: n = randint(2, 6) sage: k = randint(1, 5) sage: G = graphs.RandomGNP(n, .5) sage: m = G.size() sage: S = graphs.GeneralizedSierpinskiGraph(G, k) sage: S.order() == n**k True sage: S.size() == m*sum([n**i for i in range(k)]) True sage: G = graphs.CompleteGraph(n) sage: S = graphs.GeneralizedSierpinskiGraph(G, k) sage: S.order() == n**k True sage: S.size() == (n*(n - 1)/2)*sum([n**i for i in range(k)]) True
The positions of the vertices of the output graph are determined from the positions of the vertices of \(G\), if any:
sage: G = graphs.HouseGraph() sage: G.get_pos() is not None True sage: H = graphs.GeneralizedSierpinskiGraph(G, 2) # needs sage.symbolic sage: H.get_pos() is not None # needs sage.symbolic True sage: G = Graph([(0, 1)]) sage: G.get_pos() is not None False sage: H = graphs.GeneralizedSierpinskiGraph(G, 2) # needs sage.symbolic sage: H.get_pos() is not None # needs sage.symbolic False
- static GoethalsSeidelGraph(k, r)#
Returns the graph \(\text{Goethals-Seidel}(k,r)\).
The graph \(\text{Goethals-Seidel}(k,r)\) comes from a construction presented in Theorem 2.4 of [GS1970]. It relies on a
(v,k)-BIBDwith \(r\) blocks and ahadamard_matrix()of order \(r+1\). The result is asage.graphs.strongly_regular_db.strongly_regular_graph()on \(v(r+1)\) vertices with degree \(k=(n+r-1)/2\).It appears under this name in Andries Brouwer’s database of strongly regular graphs.
INPUT:
k,r– integers
See also
EXAMPLES:
sage: graphs.GoethalsSeidelGraph(3,3) # needs sage.combinat sage.modules Graph on 28 vertices sage: graphs.GoethalsSeidelGraph(3,3).is_strongly_regular(parameters=True) # needs sage.combinat sage.modules (28, 15, 6, 10)
- static GoldnerHararyGraph()#
Return the Goldner-Harary graph.
For more information, see the Wikipedia article Goldner%E2%80%93Harary_graph.
EXAMPLES:
The Goldner-Harary graph is named after A. Goldner and Frank Harary. It is a planar graph having 11 vertices and 27 edges:
sage: G = graphs.GoldnerHararyGraph(); G Goldner-Harary graph: Graph on 11 vertices sage: G.is_planar() True sage: G.order() 11 sage: G.size() 27
The Goldner-Harary graph is chordal with radius 2, diameter 2, and girth 3:
sage: G.is_chordal() True sage: G.radius() 2 sage: G.diameter() 2 sage: G.girth() 3
Its chromatic number is 4 and its automorphism group is isomorphic to the dihedral group \(D_6\):
sage: G.chromatic_number() 4 sage: ag = G.automorphism_group() # needs sage.groups sage: ag.is_isomorphic(DihedralGroup(6)) # needs sage.groups True
- static GolombGraph()#
Return the Golomb graph.
See the Wikipedia article Golomb_graph for more information.
EXAMPLES:
The Golomb graph is a planar and Hamiltonian graph with 10 vertices and 18 edges. It has chromatic number 4, diameter 3, radius 2 and girth 3. It can be drawn in the plane as a unit distance graph:
sage: G = graphs.GolombGraph(); G # needs sage.symbolic Golomb graph: Graph on 10 vertices sage: pos = G.get_pos() # needs sage.symbolic sage: def dist2(u, v): ....: return (u[0]-v[0])**2 + (u[1]-v[1])**2 sage: all(dist2(pos[u], pos[v]) == 1 for u, v in G.edge_iterator(labels=None)) # needs sage.symbolic True
- static GossetGraph()#
Return the Gosset graph.
The Gosset graph is the skeleton of the
Gosset_3_21()polytope. It has with 56 vertices and degree 27. For more information, see the Wikipedia article Gosset_graph.EXAMPLES:
sage: g = graphs.GossetGraph(); g Gosset Graph: Graph on 56 vertices sage: g.order(), g.size() (56, 756)
- static GrassmannGraph(q, n, input_e)#
Return the Grassmann graph with parameters \((q, n, e)\).
This builds the Grassmann graph \(J_q(n,e)\). That is, for a vector space \(V = \mathbb F(q)^n\) the output is the graph on the subspaces of dimension \(e\) where two subspaces are adjacent if their intersection has dimension \(e-1\).
This graph is distance-regular with classical parameters \((\min(e, n-e), q, q, \genfrac {[}{]} {0pt} {} {n-e+1} 1 _q -1)\)
INPUT:
q– a prime powern, e– integers withn > e+1
EXAMPLES:
sage: G = graphs.GrassmannGraph(2, 4, 2) # needs sage.modules sage.rings.finite_rings sage: G.is_distance_regular(True) # needs sage.modules sage.rings.finite_rings ([18, 8, None], [None, 1, 9])
REFERENCES:
See [BCN1989] pp. 268-272 or [VDKT2016] p. 21.
- static GrayGraph(embedding=1)#
Return the Gray graph.
See the Wikipedia article Gray_graph.
INPUT:
embedding– integer (default:1); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.GrayGraph() sage: g.order() 54 sage: g.size() 81 sage: g.girth() 8 sage: g.diameter() 6 sage: g.show(figsize=[10, 10]) # long time # needs sage.plot sage: graphs.GrayGraph(embedding=2).show(figsize=[10, 10]) # long time, needs sage.plot
- static Grid2dGraph(p, q, set_positions=True)#
Return a \(2\)-dimensional grid graph with \(p \times q\) nodes (\(p\) rows and \(q\) columns).
A 2d grid graph resembles a \(2\) dimensional grid. All inner nodes are connected to their \(4\) neighbors. Outer (non-corner) nodes are connected to their \(3\) neighbors. Corner nodes are connected to their 2 neighbors.
INPUT:
pandq– two positive integersset_positions– boolean (default:True); whether to set the position of the nodes
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, nodes are labelled in (row, column) pairs with \((0, 0)\) in the top left corner. Edges will always be horizontal and vertical - another advantage of filling the position dictionary.
EXAMPLES:
Construct and show a grid 2d graph Rows = \(5\), Columns = \(7\):
sage: g = graphs.Grid2dGraph(5,7) sage: g.show() # long time # needs sage.plot
- static GridGraph(dim_list)#
Return an \(n\)-dimensional grid graph.
INPUT:
dim_list– a list of integers representing the number of nodes toextend in each dimension
PLOTTING: When plotting, this graph will use the default spring-layout algorithm, unless a position dictionary is specified.
EXAMPLES:
sage: G = graphs.GridGraph([2,3,4]) sage: G.show() # long time # needs sage.plot
sage: C = graphs.CubeGraph(4) sage: G = graphs.GridGraph([2,2,2,2]) sage: C.show() # long time # needs sage.plot sage: G.show() # long time # needs sage.plot
- static GritsenkoGraph()#
Return SRG(65, 32, 15, 16) constructed by Gritsenko.
We took the adjacency matrix from O.Gritsenko’s [Gri2021] and extracted orbits of the automorphism group on the edges.
EXAMPLES:
sage: H = graphs.GritsenkoGraph(); H # needs sage.groups Gritsenko strongly regular graph: Graph on 65 vertices sage: H.is_strongly_regular(parameters=True) # needs sage.groups (65, 32, 15, 16)
- static GrotzschGraph()#
Return the Grötzsch graph.
The Grötzsch graph is an example of a triangle-free graph with chromatic number equal to 4. For more information, see the Wikipedia article Gr%C3%B6tzsch_graph.
EXAMPLES:
The Grötzsch graph is named after Herbert Grötzsch. It is a Hamiltonian graph with 11 vertices and 20 edges:
sage: G = graphs.GrotzschGraph(); G Grotzsch graph: Graph on 11 vertices sage: G.is_hamiltonian() # needs sage.numerical.mip True sage: G.order() 11 sage: G.size() 20
The Grötzsch graph is triangle-free and having radius 2, diameter 2, and girth 4:
sage: G.is_triangle_free() True sage: G.radius() 2 sage: G.diameter() 2 sage: G.girth() 4
Its chromatic number is 4 and its automorphism group is isomorphic to the dihedral group \(D_5\):
sage: G.chromatic_number() 4 sage: ag = G.automorphism_group() # needs sage.groups sage: ag.is_isomorphic(DihedralGroup(5)) # needs sage.groups True
- static HaemersGraph(q, hyperoval=None, hyperoval_matching=None, field=None, check_hyperoval=True)#
Return the Haemers graph obtained from \(T_2^*(q)^*\)
Let \(q\) be a power of 2. In Sect. 8.A of [BL1984] one finds a construction of a strongly regular graph with parameters \((q^2(q+2),q^2+q-1,q-2,q)\) from the graph of \(T_2^*(q)^*\), constructed by
T2starGeneralizedQuadrangleGraph(), by redefining adjacencies in the way specified by an arbitraryhyperoval_matchingof the points (i.e. partitioning into size two parts) ofhyperovaldefining \(T_2^*(q)^*\).While [BL1984] gives the construction in geometric terms, it can be formulated, and is implemented, in graph-theoretic ones, of re-adjusting the edges. Namely, \(G=T_2^*(q)^*\) has a partition into \(q+2\) independent sets \(I_k\) of size \(q^2\) each. Each vertex in \(I_j\) is adjacent to \(q\) vertices from \(I_k\). Each \(I_k\) is paired to some \(I_{k'}\), according to
hyperoval_matching. One adds edges \((s,t)\) for \(s,t \in I_k\) whenever \(s\) and \(t\) are adjacent to some \(u \in I_{k'}\), and removes all the edges between \(I_k\) and \(I_{k'}\).INPUT:
q– a power of twohyperoval_matching– ifNone(default), pair each \(i\)-th point ofhyperovalwith \((i+1)\)-th. Otherwise, specifies the pairing in the format \(((i_1,i'_1),(i_2,i'_2),...)\).hyperoval– a hyperoval defining \(T_2^*(q)^*\). IfNone(default), the classical hyperoval obtained from a conic is used. See the documentation ofT2starGeneralizedQuadrangleGraph(), for more information.field– an instance of a finite field of order \(q\), must be provided ifhyperovalis providedcheck_hyperoval– boolean (default:True); whether to checkhyperovalfor correctness or not
EXAMPLES:
using the built-in constructions:
sage: # needs sage.combinat sage.rings.finite_rings sage: g = graphs.HaemersGraph(4); g Haemers(4): Graph on 96 vertices sage: g.is_strongly_regular(parameters=True) (96, 19, 2, 4)
supplying your own hyperoval_matching:
sage: # needs sage.combinat sage.rings.finite_rings sage: g = graphs.HaemersGraph(4, hyperoval_matching=((0,5),(1,4),(2,3))); g Haemers(4): Graph on 96 vertices sage: g.is_strongly_regular(parameters=True) (96, 19, 2, 4)
- static HalfCube(n)#
Return the halved cube in \(n\) dimensions.
The graph is distance-regular with classical parameters \((\lfloor \frac n 2 \rfloor, 1, 2, 2 \lceil \frac n 2 \rceil -1)\).
INPUT:
n– integer; must be greater than 2
EXAMPLES:
sage: G = graphs.HalfCube(8) sage: G.is_distance_regular(True) ([28, 15, 6, 1, None], [None, 1, 6, 15, 28]) sage: G = graphs.HalfCube(4) sage: G.is_distance_regular(True) ([6, 1, None], [None, 1, 6])
REFERENCES:
See [BCN1989] pp. 264, 265 or [VDKT2016] p. 21. This construction can be found on Wikipedia article Halved_cube_graph#Equivalent_constructions
- static HallJankoGraph(from_string=True)#
Return the Hall-Janko graph.
For more information on the Hall-Janko graph, see the Wikipedia article Hall-Janko_graph.
The construction used to generate this graph in Sage is by a 100-point permutation representation of the Janko group \(J_2\), as described in version 3 of the ATLAS of Finite Group representations, in particular on the page ATLAS: J2 – Permutation representation on 100 points.
INPUT:
from_string– boolean (default:True); whether to build the graph from its sparse6 string or through GAP. The two methods return the same graph though doing it through GAP takes more time.
EXAMPLES:
sage: g = graphs.HallJankoGraph() sage: g.is_regular(36) True sage: g.is_vertex_transitive() # needs sage.groups True
Is it really strongly regular with parameters 14, 12?
sage: nu = set(g.neighbors(0)) sage: for v in range(1, 100): ....: if v in nu: ....: expected = 14 ....: else: ....: expected = 12 ....: nv = set(g.neighbors(v)) ....: nv.discard(0) ....: if len(nu & nv) != expected: ....: print("Something is wrong here!!!") ....: break
Some other properties that we know how to check:
sage: g.diameter() 2 sage: g.girth() 3 sage: factor(g.characteristic_polynomial()) # needs sage.libs.pari sage.modules (x - 36) * (x - 6)^36 * (x + 4)^63
- static HammingGraph(n, q, X=None)#
Returns the Hamming graph with parameters
n,qoverX.Hamming graphs are graphs over the cartesian product of n copies of
X, where \(q = |X|\), where the vertices, labelled with the corresponding tuple in \(X^n\), are connected if the Hamming distance between their labels is 1. All Hamming graphs are regular, vertex-transitive and distance-regular.Hamming graphs with parameters \((1,q)\) represent the complete graph with q vertices over the set
X.INPUT:
n– power to whichXwill be raised to provide verticesfor the Hamming graph
q– cardinality ofXX– list of labels representing the vertices of theunderlying graph the Hamming graph will be based on; if
None(or left unused), the list \([0, ... , q-1]\) will be used
OUTPUT:
G– The Hamming graph with parameters \((n,q,X)\)
EXAMPLES:
Every Hamming graph is distance-regular, regular and vertex-transitive.
sage: g = graphs.HammingGraph(3, 7) sage: g.is_distance_regular() True sage: g.is_regular() True sage: g.is_vertex_transitive() # needs sage.groups True
A Hamming graph with parameters (1,q) is isomorphic to the Complete graph with parameter q.
sage: g = graphs.HammingGraph(1, 23) sage: g.is_isomorphic(graphs.CompleteGraph(23)) True
If a parameter
qis provided which is not equal toX’s cardinality, an exception is raised.sage: X = ['a','b','c','d','e'] sage: g = graphs.HammingGraph(2, 3, X) Traceback (most recent call last): ... ValueError: q must be the cardinality of X
REFERENCES:
For a more accurate description, see the following wikipedia page: Wikipedia article Hamming_graph
- static HanoiTowerGraph(pegs, disks, labels=True, positions=True)#
Returns the graph whose vertices are the states of the Tower of Hanoi puzzle, with edges representing legal moves between states.
INPUT:
pegs- the number of pegs in the puzzle, 2 or greaterdisks- the number of disks in the puzzle, 1 or greaterlabels- default:True, ifTruethe graph contains more meaningful labels, see explanation below. For large instances, turn off labels for much faster creation of the graph.positions- default:True, ifTruethe graph contains layout information. This creates a planar layout for the case of three pegs. For large instances, turn off layout information for much faster creation of the graph.
OUTPUT:
The Tower of Hanoi puzzle has a certain number of identical pegs and a certain number of disks, each of a different radius. Initially the disks are all on a single peg, arranged in order of their radii, with the largest on the bottom.
The goal of the puzzle is to move the disks to any other peg, arranged in the same order. The one constraint is that the disks resident on any one peg must always be arranged with larger radii lower down.
The vertices of this graph represent all the possible states of this puzzle. Each state of the puzzle is a tuple with length equal to the number of disks, ordered by largest disk first. The entry of the tuple is the peg where that disk resides. Since disks on a given peg must go down in size as we go up the peg, this totally describes the state of the puzzle.
For example
(2,0,0)means the large disk is on peg 2, the medium disk is on peg 0, and the small disk is on peg 0 (and we know the small disk must be above the medium disk). We encode these tuples as integers with a base equal to the number of pegs, and low-order digits to the right.Two vertices are adjacent if we can change the puzzle from one state to the other by moving a single disk. For example,
(2,0,0)is adjacent to(2,0,1)since we can move the small disk off peg 0 and onto (the empty) peg 1. So the solution to a 3-disk puzzle (with at least two pegs) can be expressed by the shortest path between(0,0,0)and(1,1,1). For more on this representation of the graph, or its properties, see [AD2010].For greatest speed we create graphs with integer vertices, where we encode the tuples as integers with a base equal to the number of pegs, and low-order digits to the right. So for example, in a 3-peg puzzle with 5 disks, the state
(1,2,0,1,1)is encoded as \(1\ast 3^4 + 2\ast 3^3 + 0\ast 3^2 + 1\ast 3^1 + 1\ast 3^0 = 139\).For smaller graphs, the labels that are the tuples are informative, but slow down creation of the graph. Likewise computing layout information also incurs a significant speed penalty. For maximum speed, turn off labels and layout and decode the vertices explicitly as needed. The
sage.rings.integer.Integer.digits()with thepadstooption is a quick way to do this, though you may want to reverse the list that is output.See also
PLOTTING:
The layout computed when
positions = Truewill look especially good for the three-peg case, when the graph is known to be planar. Except for two small cases on 4 pegs, the graph is otherwise not planar, and likely there is a better way to layout the vertices.EXAMPLES:
A classic puzzle uses 3 pegs. We solve the 5 disk puzzle using integer labels and report the minimum number of moves required. Note that \(3^5-1\) is the state where all 5 disks are on peg 2.
sage: H = graphs.HanoiTowerGraph(3, 5, labels=False, positions=False) sage: H.distance(0, 3^5-1) 31
A slightly larger instance.
sage: H = graphs.HanoiTowerGraph(4, 6, labels=False, positions=False) sage: H.num_verts() 4096 sage: H.distance(0, 4^6-1) 17
For a small graph, labels and layout information can be useful. Here we explicitly list a solution as a list of states.
sage: H = graphs.HanoiTowerGraph(3, 3, labels=True, positions=True) sage: H.shortest_path((0,0,0), (1,1,1)) [(0, 0, 0), (0, 0, 1), (0, 2, 1), (0, 2, 2), (1, 2, 2), (1, 2, 0), (1, 1, 0), (1, 1, 1)]
Some facts about this graph with \(p\) pegs and \(d\) disks:
only automorphisms are the “obvious” ones - renumber the pegs.
chromatic number is less than or equal to \(p\)
independence number is \(p^{d-1}\)
sage: H = graphs.HanoiTowerGraph(3, 4, labels=False, positions=False) sage: H.automorphism_group().is_isomorphic(SymmetricGroup(3)) # needs sage.groups True sage: H.chromatic_number() 3 sage: len(H.independent_set()) == 3^(4-1) True
AUTHOR:
Rob Beezer, (2009-12-26), with assistance from Su Doree
- static HararyGraph(k, n)#
Returns the Harary graph on \(n\) vertices and connectivity \(k\), where \(2 \leq k < n\).
A \(k\)-connected graph \(G\) on \(n\) vertices requires the minimum degree \(\delta(G)\geq k\), so the minimum number of edges \(G\) should have is \(\lceil kn/2\rceil\). Harary graphs achieve this lower bound, that is, Harary graphs are minimal \(k\)-connected graphs on \(n\) vertices.
The construction provided uses the method CirculantGraph. For more details, see the book D. B. West, Introduction to Graph Theory, 2nd Edition, Prentice Hall, 2001, p. 150–151; or the MathWorld article on Harary graphs.
EXAMPLES:
Harary graphs \(H_{k,n}\):
sage: h = graphs.HararyGraph(5,9); h Harary graph 5, 9: Graph on 9 vertices sage: h.order() 9 sage: h.size() 23 sage: h.vertex_connectivity() # needs sage.numerical.mip 5
- static HarborthGraph()#
Return the Harborth Graph.
The Harborth graph has 104 edges and 52 vertices, and is the smallest known example of a 4-regular matchstick graph. For more information, see the Wikipedia article Harborth_graph.
EXAMPLES:
sage: g = graphs.HarborthGraph(); g Harborth Graph: Graph on 52 vertices sage: g.is_regular(4) True
- static HarriesGraph(embedding=1)#
Return the Harries Graph.
The Harries graph is a Hamiltonian 3-regular graph on 70 vertices. See the Wikipedia article Harries_graph.
The default embedding here is to emphasize the graph’s 4 orbits. This graph actually has a funny construction. The following procedure gives an idea of it, though not all the adjacencies are being properly defined.
Take two disjoint copies of a
Petersen graph. Their vertices will form an orbit of the final graph.Subdivide all the edges once, to create 15+15=30 new vertices, which together form another orbit.
Create 15 vertices, each of them linked to 2 corresponding vertices of the previous orbit, one in each of the two subdivided Petersen graphs. At the end of this step all vertices from the previous orbit have degree 3, and the only vertices of degree 2 in the graph are those that were just created.
Create 5 vertices connected only to the ones from the previous orbit so that the graph becomes 3-regular.
INPUT:
embedding– integer (default:1); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.HarriesGraph() sage: g.order() 70 sage: g.size() 105 sage: g.girth() 10 sage: g.diameter() 6 sage: g.show(figsize=[10, 10]) # long time # needs sage.plot sage: graphs.HarriesGraph(embedding=2).show(figsize=[10, 10]) # long time, needs sage.plot
- static HarriesWongGraph(embedding=1)#
Return the Harries-Wong Graph.
See the Wikipedia article Harries-Wong_graph.
About the default embedding:
The default embedding is an attempt to emphasize the graph’s 8 (!!!) different orbits. In order to understand this better, one can picture the graph as being built in the following way.
One first creates a 3-dimensional cube (8 vertices, 12 edges), whose vertices define the first orbit of the final graph.
The edges of this graph are subdivided once, to create 12 new vertices which define a second orbit.
The edges of the graph are subdivided once more, to create 24 new vertices giving a third orbit.
4 vertices are created and made adjacent to the vertices of the second orbit so that they have degree 3. These 4 vertices also define a new orbit.
In order to make the vertices from the third orbit 3-regular (they all miss one edge), one creates a binary tree on 1 + 3 + 6 + 12 vertices. The leaves of this new tree are made adjacent to the 12 vertices of the third orbit, and the graph is now 3-regular. This binary tree contributes 4 new orbits to the Harries-Wong graph.
INPUT:
embedding– integer (default:1); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.HarriesWongGraph() sage: g.order() 70 sage: g.size() 105 sage: g.girth() 10 sage: g.diameter() 6 sage: orbits = g.automorphism_group(orbits=True)[-1] # long time # needs sage.groups sage: g.show(figsize=[15, 15], partition=orbits) # long time # needs sage.groups sage.plot
Alternative embedding:
sage: graphs.HarriesWongGraph(embedding=2).show() # long time # needs networkx sage.plot
- static HeawoodGraph()#
Return a Heawood graph.
The Heawood graph is a cage graph that has 14 nodes. It is a cubic symmetric graph. (See also the Möbius-Kantor graph,
MobiusKantorGraph()). It is nonplanar and Hamiltonian. It has diameter 3, radius 3, girth 6, and chromatic number 2. It is 4-transitive but not 5-transitive. See the Wikipedia article Heawood_graph.PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the nodes are positioned in a circular layout with the first node appearing at the top, and then continuing counterclockwise.
EXAMPLES:
sage: H = graphs.HeawoodGraph() sage: H Heawood graph: Graph on 14 vertices sage: H.graph6_string() 'MhEGHC@AI?_PC@_G_' sage: (graphs.HeawoodGraph()).show() # long time # needs sage.plot
- static HermitianFormsGraph(n, r)#
Return the Hermitian forms graph with the given parameters.
We build a graph whose vertices are all
n``x``nHermitian matrices overGF(r^2). Two vertices are adjacent if the difference of the two vertices has rank 1.This graph is distance-regular with classical parameters \((n, - r, - r - 1, - (- r)^d - 1)\).
INPUT:
n– integerr– a prime power
EXAMPLES:
sage: # needs sage.modules sage.rings.finite_rings sage: G = graphs.HermitianFormsGraph(2, 2) sage: G.is_distance_regular(True) ([5, 4, None], [None, 1, 2]) sage: G = graphs.HermitianFormsGraph(3, 3) # not tested (2 min) sage: G.order() # not tested (bacuase of the above) 19683
REFERENCES:
See [BCN1989] p. 285 or [VDKT2016] p. 22.
- static HerschelGraph()#
Return the Herschel graph.
For more information, see the Wikipedia article Herschel_graph.
EXAMPLES:
The Herschel graph is named after Alexander Stewart Herschel. It is a planar, bipartite graph with 11 vertices and 18 edges:
sage: G = graphs.HerschelGraph(); G Herschel graph: Graph on 11 vertices sage: G.is_planar() True sage: G.is_bipartite() True sage: G.order() 11 sage: G.size() 18
The Herschel graph is a perfect graph with radius 3, diameter 4, and girth 4:
sage: G.is_perfect() True sage: G.radius() 3 sage: G.diameter() 4 sage: G.girth() 4
Its chromatic number is 2 and its automorphism group is isomorphic to the dihedral group \(D_6\):
sage: G.chromatic_number() 2 sage: ag = G.automorphism_group() # needs sage.groups sage: ag.is_isomorphic(DihedralGroup(6)) # needs sage.groups True
- static HexahedralGraph()#
Return a hexahedral graph (with 8 nodes).
A regular hexahedron is a 6-sided cube. The hexahedral graph corresponds to the connectivity of the vertices of the hexahedron. This graph is equivalent to a 3-cube.
PLOTTING: The Hexahedral graph should be viewed in 3 dimensions. We choose to use a planar embedding of the graph. We hope to add rotatable, 3-dimensional viewing in the future. In such a case, a argument will be added to select the desired layout.
EXAMPLES:
Construct and show a Hexahedral graph:
sage: g = graphs.HexahedralGraph() sage: g.show() # long time # needs sage.plot
Create several hexahedral graphs in a Sage graphics array. They will be drawn differently due to the use of the spring-layout algorithm:
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.HexahedralGraph() ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static HigmanSimsGraph(relabel=True)#
Return the Higman-Sims graph.
The Higman-Sims graph is a remarkable strongly regular graph of degree 22 on 100 vertices. For example, it can be split into two sets of 50 vertices each, so that each half induces a subgraph isomorphic to the Hoffman-Singleton graph (
HoffmanSingletonGraph()). This can be done in 352 ways (see Higman-Sims graph by Andries E. Brouwer, accessed 24 October 2009.)Its most famous property is that the automorphism group has an index 2 subgroup which is one of the 26 sporadic groups [HS1968].
The construction used here follows [Haf2004].
See also the Wikipedia article Higman–Sims_graph.
INPUT:
relabel– boolean (default:True); whether to relabel the vertices with consecutive integers. IfFalsethe labels are strings that are three digits long. “xyz” means the vertex is in group \(x\) (zero through three), pentagon or pentagram \(y\) (zero through four), and is vertex \(z\) (zero through four) of that pentagon or pentagram. See [Haf2004] for more.
OUTPUT:
The Higman-Sims graph.
EXAMPLES:
A split into the first 50 and last 50 vertices will induce two copies of the Hoffman-Singleton graph, and we illustrate another such split, which is obvious based on the construction used:
sage: H = graphs.HigmanSimsGraph() sage: A = H.subgraph(range(0,50)) sage: B = H.subgraph(range(50,100)) sage: K = graphs.HoffmanSingletonGraph() sage: K.is_isomorphic(A) and K.is_isomorphic(B) True sage: C = H.subgraph(range(25,75)) sage: D = H.subgraph(list(range(0,25))+list(range(75,100))) sage: K.is_isomorphic(C) and K.is_isomorphic(D) True
The automorphism group contains only one nontrivial proper normal subgroup, which is of index 2 and is simple. It is known as the Higman-Sims group:
sage: H = graphs.HigmanSimsGraph() sage: G = H.automorphism_group() # needs sage.groups sage: g = G.order(); g # needs sage.groups 88704000 sage: K = G.normal_subgroups()[1] # needs sage.groups sage: K.is_simple() # needs sage.groups True sage: g//K.order() # needs sage.groups 2
AUTHOR:
Rob Beezer (2009-10-24)
- static HoffmanGraph()#
Return the Hoffman Graph.
See the Wikipedia article Hoffman_graph.
EXAMPLES:
sage: g = graphs.HoffmanGraph() sage: g.is_bipartite() True sage: g.is_hamiltonian() # long time # needs sage.numerical.mip True sage: g.radius() 3 sage: g.diameter() 4 sage: g.automorphism_group().cardinality() # needs sage.groups 48
- static HoffmanSingletonGraph()#
Return the Hoffman-Singleton graph.
The Hoffman-Singleton graph is the Moore graph of degree 7, diameter 2 and girth 5. The Hoffman-Singleton theorem states that any Moore graph with girth 5 must have degree 2, 3, 7 or 57. The first three respectively are the pentagon, the Petersen graph, and the Hoffman-Singleton graph. The existence of a Moore graph with girth 5 and degree 57 is still open.
A Moore graph is a graph with diameter \(d\) and girth \(2d + 1\). This implies that the graph is regular, and distance regular.
For more details, see [GR2001] and the Wikipedia article Hoffman–Singleton_graph.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. A novel algorithm written by Tom Boothby gives a random layout which is pleasing to the eye.
EXAMPLES:
sage: HS = graphs.HoffmanSingletonGraph() sage: Set(HS.degree()) {7} sage: HS.girth() 5 sage: HS.diameter() 2 sage: HS.num_verts() 50
Note that you get a different layout each time you create the graph.
sage: HS.layout()[1] # random (-0.844..., 0.535...) sage: HS = graphs.HoffmanSingletonGraph() sage: HS.layout()[1] # random (-0.904..., 0.425...)
- static HoltGraph()#
Return the Holt graph (also called the Doyle graph).
See the Wikipedia article Holt_graph.
EXAMPLES:
sage: g = graphs.HoltGraph();g Holt graph: Graph on 27 vertices sage: g.is_regular() True sage: g.is_vertex_transitive() # needs sage.groups True sage: g.chromatic_number() 3 sage: g.is_hamiltonian() # long time # needs sage.numerical.mip True sage: g.radius() 3 sage: g.diameter() 3 sage: g.girth() 5 sage: g.automorphism_group().cardinality() # needs sage.groups 54
- static HortonGraph()#
Return the Horton Graph.
The Horton graph is a cubic 3-connected non-hamiltonian graph. For more information, see the Wikipedia article Horton_graph.
EXAMPLES:
sage: # needs networkx sage: g = graphs.HortonGraph() sage: g.order() 96 sage: g.size() 144 sage: g.radius() 10 sage: g.diameter() 10 sage: g.girth() 6 sage: g.automorphism_group().cardinality() 96 sage: g.chromatic_number() 2 sage: g.is_hamiltonian() # not tested (veeeery long) # needs sage.numerical.mip False
- static HouseGraph()#
Return a house graph with 5 nodes.
A house graph is named for its shape. It is a triangle (roof) over a square (walls).
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the house graph is drawn with the first node in the lower-left corner of the house, the second in the lower-right corner of the house. The third node is in the upper-left corner connecting the roof to the wall, and the fourth is in the upper-right corner connecting the roof to the wall. The fifth node is the top of the roof, connected only to the third and fourth.
EXAMPLES:
Construct and show a house graph:
sage: g = graphs.HouseGraph() sage: g.show() # long time # needs sage.plot
- static HouseXGraph()#
Return a house X graph with 5 nodes.
A house X graph is a house graph with two additional edges. The upper-right corner is connected to the lower-left. And the upper-left corner is connected to the lower-right.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the house X graph is drawn with the first node in the lower-left corner of the house, the second in the lower-right corner of the house. The third node is in the upper-left corner connecting the roof to the wall, and the fourth is in the upper-right corner connecting the roof to the wall. The fifth node is the top of the roof, connected only to the third and fourth.
EXAMPLES:
Construct and show a house X graph:
sage: g = graphs.HouseXGraph() sage: g.show() # long time # needs sage.plot
- static HyperStarGraph(n, k)#
Return the hyper-star graph \(HS(n, k)\).
The vertices of the hyper-star graph are the set of binary strings of length \(n\) which contain \(k\) 1s. Two vertices, \(u\) and \(v\), are adjacent only if \(u\) can be obtained from \(v\) by swapping the first bit with a different symbol in another position. For instance, vertex
'011100'of \(HS(6, 3)\) is adjacent to vertices'101100','110100'and'111000'. See [LKOL2002] for more details.INPUT:
n– non-negative integer; length of the binary stringsk– non-negative integer; number of 1s per binary string
EXAMPLES:
sage: g = graphs.HyperStarGraph(6,3) sage: sorted(g.neighbors('011100')) ['101100', '110100', '111000'] sage: g.plot() # long time # needs sage.plot Graphics object consisting of 51 graphics primitives
AUTHORS:
Michael Yurko (2009-09-01)
- static IGraph(n, j, k)#
Return an I-graph with \(2n\) nodes.
The I-Graph family as been proposed in [BCMS1988] as a generalization of the generalized Petersen graphs. The variables \(n\), \(j\), \(k\) are integers such that \(n > 2\) and \(0 < j, k \leq \lfloor (n - 1) / 2 \rfloor\). When \(j = 1\) the resulting graph is isomorphic to the generalized Petersen graph with the same \(n\) and \(k\).
INPUT:
n– the number of nodes is \(2 * n\)j– integer such that \(0 < j \leq \lfloor (n-1) / 2 \rfloor\) determining how outer vertices are connectedk– integer such that \(0 < k \leq \lfloor (n-1) / 2 \rfloor\) determining how inner vertices are connected
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the I-graphs are displayed as an inner and outer cycle pair, with the first n nodes drawn on the outer circle. The first (0) node is drawn at the top of the outer-circle, moving counterclockwise after that. The inner circle is drawn with the (n)th node at the top, then counterclockwise as well.
EXAMPLES:
When \(j = 1\) the resulting graph will be isomorphic to a generalized Petersen graph:
sage: g = graphs.IGraph(7,1,2) sage: g2 = graphs.GeneralizedPetersenGraph(7,2) sage: g.is_isomorphic(g2) True
The IGraph with parameters \((n, j, k)\) is isomorphic to the IGraph with parameters \((n, k, j)\):
sage: g = graphs.IGraph(7, 2, 3) sage: h = graphs.IGraph(7, 3, 2) sage: g.is_isomorphic(h) True
- static IcosahedralGraph()#
Return an Icosahedral graph (with 12 nodes).
The regular icosahedron is a 20-sided triangular polyhedron. The icosahedral graph corresponds to the connectivity of the vertices of the icosahedron. It is dual to the dodecahedral graph. The icosahedron is symmetric, so the spring-layout algorithm will be very effective for display.
PLOTTING: The Icosahedral graph should be viewed in 3 dimensions. We choose to use a planar embedding of the graph. We hope to add rotatable, 3-dimensional viewing in the future. In such a case, a argument will be added to select the desired layout.
EXAMPLES:
Construct and show an Octahedral graph:
sage: g = graphs.IcosahedralGraph() sage: g.show() # long time # needs sage.plot
Create several icosahedral graphs in a Sage graphics array. They will be drawn differently due to the use of the spring-layout algorithm:
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.IcosahedralGraph() ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static IntersectionGraph(S)#
Return the intersection graph of the family \(S\)
The intersection graph of a family \(S\) is a graph \(G\) with \(V(G)=S\) such that two elements \(s_1,s_2\in S\) are adjacent in \(G\) if and only if \(s_1\cap s_2\neq \emptyset\).
INPUT:
S– a list of sets/tuples/iterablesNote
The elements of \(S\) must be finite, hashable, and the elements of any \(s\in S\) must be hashable too.
EXAMPLES:
sage: graphs.IntersectionGraph([(1,2,3),(3,4,5),(5,6,7)]) Intersection Graph: Graph on 3 vertices
- static IntervalGraph(intervals, points_ordered=False)#
Return the graph corresponding to the given intervals.
An interval graph is built from a list \((a_i,b_i)_{1\leq i \leq n}\) of intervals : to each interval of the list is associated one vertex, two vertices being adjacent if the two corresponding (closed) intervals intersect.
INPUT:
intervals– the list of pairs \((a_i,b_i)\) defining the graph.points_ordered– states whether every interval \((a_i,b_i)\) of \(intervals\) satisfies \(a_i<b_i\). If satisfied then settingpoints_orderedtoTruewill speed up the creation of the graph.
Note
The vertices are named 0, 1, 2, and so on. The intervals used to create the graph are saved with the graph and can be recovered using
get_vertex()orget_vertices().
EXAMPLES:
The following line creates the sequence of intervals \((i, i+2)\) for i in \([0, ..., 8]\):
sage: intervals = [(i,i+2) for i in range(9)]
In the corresponding graph
sage: g = graphs.IntervalGraph(intervals) sage: g.get_vertex(3) (3, 5) sage: neigh = g.neighbors(3) sage: for v in neigh: print(g.get_vertex(v)) (1, 3) (2, 4) (4, 6) (5, 7)
The is_interval() method verifies that this graph is an interval graph.
sage: g.is_interval() True
The intervals in the list need not be distinct.
sage: intervals = [ (1,2), (1,2), (1,2), (2,3), (3,4) ] sage: g = graphs.IntervalGraph(intervals,True) sage: g.clique_maximum() [0, 1, 2, 3] sage: g.get_vertices() {0: (1, 2), 1: (1, 2), 2: (1, 2), 3: (2, 3), 4: (3, 4)}
The endpoints of the intervals are not ordered we get the same graph (except for the vertex labels).
sage: rev_intervals = [ (2,1), (2,1), (2,1), (3,2), (4,3) ] sage: h = graphs.IntervalGraph(rev_intervals,False) sage: h.get_vertices() {0: (2, 1), 1: (2, 1), 2: (2, 1), 3: (3, 2), 4: (4, 3)} sage: g.edges(sort=True) == h.edges(sort=True) True
- static IoninKharaghani765Graph()#
Return a \((765, 192, 48, 48)\)-strongly regular graph.
Existence of a strongly regular graph with these parameters was claimed in [IK2003]. Implementing the construction in the latter did not work, however. This function implements the following instructions, shared by Yury Ionin and Hadi Kharaghani.
Let \(A\) be the affine plane over the field \(GF(3)=\{-1,0,1\}\). Let
\[\begin{split}\phi_1(x,y) &= x\\ \phi_2(x,y) &= y\\ \phi_3(x,y) &= x+y\\ \phi_4(x,y) &= x-y\\\end{split}\]For \(i=1,2,3,4\) and \(j\in GF(3)\), let \(L_{i,j}\) be the line in \(A\) defined by \(\phi_i(x,y)=j\). Let \(\mathcal M\) be the set of all 12 lines \(L_{i,j}\), plus the empty set. Let \(\pi\) be the permutation defined on \(\mathcal M\) by \(\pi(L_{i,j}) = L_{i,j+1}\) and \(\pi(\emptyset) = \emptyset\), so that \(\pi\) has three orbits of cardinality 3 and one of cardinality 1.
Let \(A=(p_1,...,p_9)\) with \(p_1=(-1,1)\), \(p_2=(-1,0)\), \(p_3=(-1,1)\), \(p_4=(0,-1)\), \(p_5=(0,0)\), \(p_6=(0,1)\), \(p_7=(1,-1)\), \(p_8=(1,0)\), \(p_9=(1,1)\). Note that \(p_i+p_{10-i}=(0,0)\). For any subset \(X\) of \(A\), let \(M(X)\) be the \((0,1)\)-matrix of order 9 whose \((i,j)\)-entry equals 1 if and only if \(p_{10-i}-p_j\in X\). Note that \(M\) is a symmetric matrix.
An \(MF\)-tuple is an ordered quintuple \((X_1, X_2, X_3, X_4, X_5)\) of subsets of \(A\), of which one is the empty set and the other four are pairwise non-parallel lines. Such a quintuple generates the following block matrix:
\[\begin{split}N(X_1, X_2, X_3, X_4, X_5) = \left( \begin{array}{ccccc} M(X_1) & M(X_2) & M(X_3) & M(X_4) & M(X_5)\\ M(X_2) & M(X_3) & M(X_4) & M(X_5) & M(X_1)\\ M(X_3) & M(X_4) & M(X_5) & M(X_1) & M(X_2)\\ M(X_4) & M(X_5) & M(X_1) & M(X_2) & M(X_3)\\ M(X_5) & M(X_1) & M(X_2) & M(X_3) & M(X_4) \end{array}\right)\end{split}\]Observe that if \((X_1, X_2, X_3, X_4, X_5)\) is an \(MF\)-tuple, then \(N(X_1, X_2, X_3, X_4, X_5)\) is the symmetric incidence matrix of a symmetric \((45, 12, 3)\)-design.
Let \(\mathcal F\) be the set of all \(MF\)-tuples and let \(\sigma\) be the following permutation of \(\mathcal F\):
\[\begin{split}\sigma(X_1, X_2, X_3, X_4, X_5) & = (X_2, X_3, X_4, X_5, X_1)\\ \pi(X_1, X_2, X_3, X_4, X_5) & = (\pi(X_1), \pi(X_2), \pi(X_3), \pi(X_4), \pi(X_5))\\\end{split}\]Observe that \(\sigma\) and \(\pi\) commute, and generate a (cyclic) group \(G\) of order 15. We will from now on identify \(G\) with the (cyclic) multiplicative group of the field \(GF(16)\) equal to \(\{\omega^0,...,\omega^{14}\}\). Let \(W=[w_{ij}]\) be the following matrix of order 17 over \(GF(16)=\{a_1,...,a_16\}\):
\[\begin{split}w_{ij}=\left\{\begin{array}{ll} a_i+a_j & \text{if }1\leq i\leq 16, 1\leq j\leq 16,\\ 1 & \text{if }i=17, j\neq 17,\\ 1 & \text{if }i\neq 17, j= 17,\\ 0 & \text{if }i=j=17 \end{array}\right.\end{split}\]The diagonal entries of \(W\) are equal to 0, each off-diagonal entry can be represented as \(\omega^k\) with \(0\leq k\leq 14\). Matrix \(W\) is a symmetric \(BGW(17,16,15; G)\).
Fix an \(MF\)-tuple \((X_1, X_2, X_3, X_4, X_5)\) and let \(S\) be the block matrix obtained from \(W\) by replacing every diagonal entry of \(W\) by the zero matrix of order 45, and every off-diagonal entry \(\omega^k\) by the matrix \(N(\sigma^k(X_1, X_2, X_3, X_4, X_5))\) (through the association of \(\omega^k\) with an element of \(G\)). Then \(S\) is a symmetric incidence matrix of a symmetric \((765, 192, 48)\)-design with zero diagonal, and therefore \(S\) is an adjacency matrix of a strongly regular graph with parameters \((765, 192, 48, 48)\).
EXAMPLES:
sage: g = graphs.IoninKharaghani765Graph(); g # needs sage.modules sage.rings.finite_rings Ionin-Kharaghani: Graph on 765 vertices
Todo
An update to [IK2003] meant to fix the problem encountered became available 2016/02/24, see http://www.cs.uleth.ca/~hadi/research/IoninKharaghani.pdf
- static IvanovIvanovFaradjevGraph()#
Return the IvanovIvanovFaradjev graph.
The graph is distance-transitive with automorphism group \(3.M_{22}\).
EXAMPLES:
sage: G = graphs.IvanovIvanovFaradjevGraph() # optional - internet gap_package_atlasrep sage: G.is_distance_regular(True) # optional - internet gap_package_atlasrep ([7, 6, 4, 4, 4, 1, 1, 1, None], [None, 1, 1, 1, 2, 4, 4, 6, 7])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 369.
- static J2Graph()#
Return the distance-transitive graph with automorphism group \(J_2\).
EXAMPLES:
sage: G = graphs.J2Graph() # optional - internet gap_package_atlasrep sage: G.is_distance_regular(True) # optional - internet gap_package_atlasrep ([10, 8, 8, 2, None], [None, 1, 1, 4, 5])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 408.
- static JankoKharaghaniGraph(v)#
Return a \((936, 375, 150, 150)\)-srg or a \((1800, 1029, 588, 588)\)-srg.
This functions returns a strongly regular graph for the two sets of parameters shown to be realizable in [JK2002]. The paper also uses a construction from [GM1987].
INPUT:
v– integer; one of 936 or 1800
EXAMPLES:
sage: g = graphs.JankoKharaghaniGraph(936) # long time # needs sage.libs.pari sage: g.is_strongly_regular(parameters=True) # long time # needs sage.libs.pari (936, 375, 150, 150) sage: g = graphs.JankoKharaghaniGraph(1800) # not tested (30s) sage: g.is_strongly_regular(parameters=True) # not tested (30s) (1800, 1029, 588, 588)
- static JankoKharaghaniTonchevGraph()#
Return a \((324,153,72,72)\)-strongly regular graph from [JKT2001].
Build the graph using the description given in [JKT2001], taking sets B1 and B163 in the text as adjacencies of vertices 1 and 163, respectively, and taking the edge orbits of the group \(G\) provided.
EXAMPLES:
sage: Gamma = graphs.JankoKharaghaniTonchevGraph() # long time # needs sage.libs.gap sage: Gamma.is_strongly_regular(parameters=True) # long time # needs sage.libs.gap (324, 153, 72, 72)
- static JohnsonGraph(n, k)#
Returns the Johnson graph with parameters \(n, k\).
Johnson graphs are a special class of undirected graphs defined from systems of sets. The vertices of the Johnson graph \(J(n,k)\) are the \(k\)-element subsets of an \(n\)-element set; two vertices are adjacent when they meet in a \((k-1)\)-element set. See the Wikipedia article Johnson_graph for more information.
EXAMPLES:
The Johnson graph is a Hamiltonian graph:
sage: g = graphs.JohnsonGraph(7, 3) sage: g.is_hamiltonian() # needs sage.numerical.mip True
Every Johnson graph is vertex transitive:
sage: g = graphs.JohnsonGraph(6, 4) sage: g.is_vertex_transitive() # needs sage.groups True
The complement of the Johnson graph \(J(n,2)\) is isomorphic to the Kneser Graph \(K(n,2)\). In particular the complement of \(J(5,2)\) is isomorphic to the Petersen graph.:
sage: g = graphs.JohnsonGraph(5,2) sage: g.complement().is_isomorphic(graphs.PetersenGraph()) True
- static KingGraph(dim_list, radius=None, relabel=False)#
Return the \(d\)-dimensional King Graph with prescribed dimensions.
The 2-dimensional King Graph of parameters \(n\) and \(m\) is a graph with \(nm\) vertices in which each vertex represents a square in an \(n \times m\) chessboard, and each edge corresponds to a legal move by a king.
The d-dimensional King Graph with \(d >= 2\) has for vertex set the cells of a d-dimensional grid with prescribed dimensions, and each edge corresponds to a legal move by a king in either one or two dimensions.
All 2-dimensional King Graphs are Hamiltonian, biconnected, and have chromatic number 4 as soon as both dimensions are larger or equal to 2.
INPUT:
dim_list– iterable (list, set, dict); provides the dimensions \(n_1, n_2, \ldots, n_d\), with \(n_i \geq 1\), of the chessboardradius– integer (default:None); by setting the radius to a positive integer, one may increase the power of the king to at leastradiussteps. When the radius equals the higher size of the dimensions, the resulting graph is a Queen Graph.relabel– boolean (default:False); indicates whether the vertices must be relabeled as integers
EXAMPLES:
The \((2,2)\)-King Graph is isomorphic to the complete graph on 4 vertices:
sage: G = graphs.QueenGraph( [2, 2] ) sage: G.is_isomorphic( graphs.CompleteGraph(4) ) True
The King Graph with large enough radius is isomorphic to a Queen Graph:
sage: G = graphs.KingGraph( [5, 4], radius=5 ) sage: H = graphs.QueenGraph( [4, 5] ) sage: G.is_isomorphic( H ) True
Also True in higher dimensions:
sage: G = graphs.KingGraph( [2, 5, 4], radius=5 ) sage: H = graphs.QueenGraph( [4, 5, 2] ) sage: G.is_isomorphic( H ) True
- static KittellGraph()#
Return the Kittell Graph.
For more information, see the Wolfram page about the Kittel Graph.
EXAMPLES:
sage: g = graphs.KittellGraph() sage: g.order() 23 sage: g.size() 63 sage: g.radius() 3 sage: g.diameter() 4 sage: g.girth() 3 sage: g.chromatic_number() 4
- static Klein3RegularGraph()#
Return the Klein 3-regular graph.
The cubic Klein graph has 56 vertices and can be embedded on a surface of genus 3. It is the dual of
Klein7RegularGraph(). For more information, see the Wikipedia article Klein_graphs.EXAMPLES:
sage: g = graphs.Klein3RegularGraph(); g Klein 3-regular Graph: Graph on 56 vertices sage: g.order(), g.size() (56, 84) sage: g.girth() 7 sage: g.automorphism_group().cardinality() # needs sage.groups 336 sage: g.chromatic_number() 3
- static Klein7RegularGraph()#
Return the Klein 7-regular graph.
The 7-valent Klein graph has 24 vertices and can be embedded on a surface of genus 3. It is the dual of
Klein3RegularGraph(). For more information, see the Wikipedia article Klein_graphs.EXAMPLES:
sage: g = graphs.Klein7RegularGraph(); g Klein 7-regular Graph: Graph on 24 vertices sage: g.order(), g.size() (24, 84) sage: g.girth() 3 sage: g.automorphism_group().cardinality() # needs sage.groups 336 sage: g.chromatic_number() 4
- static KneserGraph(n, k)#
Returns the Kneser Graph with parameters \(n, k\).
The Kneser Graph with parameters \(n,k\) is the graph whose vertices are the \(k\)-subsets of \([0,1,\dots,n-1]\), and such that two vertices are adjacent if their corresponding sets are disjoint.
For example, the Petersen Graph can be defined as the Kneser Graph with parameters \(5,2\).
EXAMPLES:
sage: KG = graphs.KneserGraph(5,2) sage: sorted(KG.vertex_iterator(), key=str) [{1, 2}, {1, 3}, {1, 4}, {1, 5}, {2, 3}, {2, 4}, {2, 5}, {3, 4}, {3, 5}, {4, 5}] sage: P = graphs.PetersenGraph() sage: P.is_isomorphic(KG) True
- static KnightGraph(dim_list, one=1, two=2, relabel=False)#
Return the d-dimensional Knight Graph with prescribed dimensions.
The 2-dimensional Knight Graph of parameters \(n\) and \(m\) is a graph with \(nm\) vertices in which each vertex represents a square in an \(n \times m\) chessboard, and each edge corresponds to a legal move by a knight.
The d-dimensional Knight Graph with \(d >= 2\) has for vertex set the cells of a d-dimensional grid with prescribed dimensions, and each edge corresponds to a legal move by a knight in any pairs of dimensions.
The \((n,n)\)-Knight Graph is Hamiltonian for even \(n > 4\).
INPUT:
dim_list– iterable (list, set, dict); provides the dimensions \(n_1, n_2, \ldots, n_d\), with \(n_i \geq 1\), of the chessboardone– integer (default:1); indicates the number of steps in the first dimensiontwo– integer (default:2); indicates the number of steps in the second dimensionrelabel– boolean (default:False); indicates whether the vertices must be relabeled as integers
EXAMPLES:
The \((3,3)\)-Knight Graph has an isolated vertex:
sage: G = graphs.KnightGraph( [3, 3] ) sage: G.degree( (1,1) ) 0
The \((3,3)\)-Knight Graph minus vertex (1,1) is a cycle of order 8:
sage: G = graphs.KnightGraph( [3, 3] ) sage: G.delete_vertex( (1,1) ) sage: G.is_isomorphic( graphs.CycleGraph(8) ) True
The \((6,6)\)-Knight Graph is Hamiltonian:
sage: G = graphs.KnightGraph( [6, 6] ) sage: G.is_hamiltonian() # needs sage.numerical.mip True
- static KrackhardtKiteGraph()#
Return a Krackhardt kite graph with 10 nodes.
The Krackhardt kite graph was originally developed by David Krackhardt for the purpose of studying social networks (see [Kre2002] and the Wikipedia article Krackhardt_kite_graph). It is used to show the distinction between degree centrality, betweenness centrality, and closeness centrality. For more information read the plotting section below in conjunction with the example.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the graph is drawn left to right, in top to bottom row sequence of [2, 3, 2, 1, 1, 1] nodes on each row. This places the fourth node (3) in the center of the kite, with the highest degree. But the fourth node only connects nodes that are otherwise connected, or those in its clique (i.e.: Degree Centrality). The eighth (7) node is where the kite meets the tail. It has degree = 3, less than the average, but is the only connection between the kite and tail (i.e.: Betweenness Centrality). The sixth and seventh nodes (5 and 6) are drawn in the third row and have degree = 5. These nodes have the shortest path to all other nodes in the graph (i.e.: Closeness Centrality). Please execute the example for visualization.
EXAMPLES:
Construct and show a Krackhardt kite graph
sage: g = graphs.KrackhardtKiteGraph() sage: g.show() # long time # needs sage.plot
- static LCFGraph(n, shift_list, repeats)#
Return the cubic graph specified in LCF notation.
LCF (Lederberg-Coxeter-Fruchte) notation is a concise way of describing cubic Hamiltonian graphs. The way a graph is constructed is as follows. Since there is a Hamiltonian cycle, we first create a cycle on n nodes. The variable shift_list = [s_0, s_1, …, s_k-1] describes edges to be created by the following scheme: for each i, connect vertex i to vertex (i + s_i). Then, repeats specifies the number of times to repeat this process, where on the jth repeat we connect vertex (i + j*len(shift_list)) to vertex ( i + j*len(shift_list) + s_i).
INPUT:
n– the number of nodes.shift_list– a list of integer shifts mod \(n\).repeats– the number of times to repeat the process.
EXAMPLES:
sage: G = graphs.LCFGraph(4, [2,-2], 2) # needs networkx sage: G.is_isomorphic(graphs.TetrahedralGraph()) # needs networkx True
sage: G = graphs.LCFGraph(20, [10,7,4,-4,-7,10,-4,7,-7,4], 2) # needs networkx sage: G.is_isomorphic(graphs.DodecahedralGraph()) # needs networkx True
sage: G = graphs.LCFGraph(14, [5,-5], 7) # needs networkx sage: G.is_isomorphic(graphs.HeawoodGraph()) # needs networkx True
The largest cubic nonplanar graph of diameter three:
sage: # needs networkx sage: G = graphs.LCFGraph(20, [-10,-7,-5,4,7,-10,-7,-4,5,7, ....: -10,-7,6,-5,7,-10,-7,5,-6,7], 1) sage: G.degree() [3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3] sage: G.diameter() 3 sage: G.show() # long time # needs sage.plot
PLOTTING: LCF Graphs are plotted as an n-cycle with edges in the middle, as described above.
REFERENCES:
[1] Frucht, R. “A Canonical Representation of Trivalent Hamiltonian Graphs.” J. Graph Th. 1, 45-60, 1976.
[2] Grunbaum, B. Convex Polytope es. New York: Wiley, pp. 362-364, 1967.
[3] Lederberg, J. ‘DENDRAL-64: A System for Computer Construction, Enumeration and Notation of Organic Molecules as Tree Structures and Cyclic Graphs. Part II. Topology of Cyclic Graphs.’ Interim Report to the National Aeronautics and Space Administration. Grant NsG 81-60. December 15, 1965. http://profiles.nlm.nih.gov/BB/A/B/I/U/_/bbabiu.pdf.
- static LadderGraph(n)#
Return a ladder graph with \(2 * n\) nodes.
A ladder graph is a basic structure that is typically displayed as a ladder, i.e.: two parallel path graphs connected at each corresponding node pair.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each ladder graph will be displayed horizontally, with the first n nodes displayed left to right on the top horizontal line.
EXAMPLES:
Construct and show a ladder graph with 14 nodes:
sage: g = graphs.LadderGraph(7) sage: g.show() # long time # needs sage.plot
Create several ladder graphs in a Sage graphics array:
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.LadderGraph(i+2) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static LargeWittGraph()#
Return the large Witt graph.
This is a distance-regular graph with intersection array \([30,28,24;1,3,15]\).
EXAMPLES:
sage: g = graphs.LargeWittGraph() # needs sage.libs.pari sage.modules sage: g.is_distance_regular(True) # needs sage.libs.pari sage.modules ([30, 28, 24, None], [None, 1, 3, 15])
REFERENCES:
A description of this graph can be found in [BCN1989] p. 366. This construction is taken from http://mathworld.wolfram.com/LargeWittGraph.html
- static LeonardGraph()#
Return the Leonard graph.
The graph is distance-regular with intersection array \([12, 11, 10, 7; 1, 2, 5, 12]\).
EXAMPLES:
sage: G = graphs.LeonardGraph() # needs sage.combinat sage.modules sage: G.is_distance_regular(True) # needs sage.combinat sage.modules ([12, 11, 10, 7, None], [None, 1, 2, 5, 12])
REFERENCES:
For a description of this graph see [BCN1989] p. 371.
- static LivingstoneGraph()#
Return the Livingstone Graph.
The Livingstone graph is a distance-transitive graph on 266 vertices whose automorphism group is the
J1 group. For more information, see the Wikipedia article Livingstone_graph.EXAMPLES:
sage: # optional - internet sage: g = graphs.LivingstoneGraph() sage: g.order() 266 sage: g.size() 1463 sage: g.girth() 5 sage: g.is_vertex_transitive() True sage: g.is_distance_regular() True
- static LjubljanaGraph(embedding=1)#
Return the Ljubljana Graph.
The Ljubljana graph is a bipartite 3-regular graph on 112 vertices and 168 edges. It is not vertex-transitive as it has two orbits which are also independent sets of size 56. See the Wikipedia article Ljubljana_graph.
The default embedding is obtained from the Heawood graph.
INPUT:
embedding– integer (default:1); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.LjubljanaGraph() sage: g.order() 112 sage: g.size() 168 sage: g.girth() 10 sage: g.diameter() 8 sage: g.show(figsize=[10, 10]) # long time # needs sage.plot sage: graphs.LjubljanaGraph(embedding=2).show(figsize=[10, 10]) # long time, needs sage.plot
- static LocalMcLaughlinGraph()#
Return the local McLaughlin graph.
The local McLaughlin graph is a strongly regular graph with parameters \((162,56,10,24)\). It can be obtained from
McLaughlinGraph()by considering the stabilizer of a point: one of its orbits has cardinality 162.EXAMPLES:
sage: g = graphs.LocalMcLaughlinGraph(); g # long time, optional - gap_package_design Local McLaughlin Graph: Graph on 162 vertices sage: g.is_strongly_regular(parameters=True) # long time, optional - gap_package_design (162, 56, 10, 24)
- static LollipopGraph(n1, n2)#
Returns a lollipop graph with n1+n2 nodes.
A lollipop graph is a path graph (order n2) connected to a complete graph (order n1). (A barbell graph minus one of the bells).
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the complete graph will be drawn in the lower-left corner with the (n1)th node at a 45 degree angle above the right horizontal center of the complete graph, leading directly into the path graph.
EXAMPLES:
Construct and show a lollipop graph Candy = 13, Stick = 4:
sage: g = graphs.LollipopGraph(13,4); g Lollipop graph: Graph on 17 vertices sage: g.show() # long time # needs sage.plot
- static M22Graph()#
Return the M22 graph.
The \(M_{22}\) graph is the unique strongly regular graph with parameters \(v = 77, k = 16, \lambda = 0, \mu = 4\).
For more information on the \(M_{22}\) graph, see https://www.win.tue.nl/~aeb/graphs/M22.html.
EXAMPLES:
sage: # needs sage.groups sage: g = graphs.M22Graph() sage: g.order() 77 sage: g.size() 616 sage: g.is_strongly_regular(parameters=True) (77, 16, 0, 4)
- static MarkstroemGraph()#
Return the Markström Graph.
The Markström Graph is a cubic planar graph with no cycles of length 4 nor 8, but containing cycles of length 16. For more information, see the Wolfram page about the Markström Graph.
EXAMPLES:
sage: g = graphs.MarkstroemGraph() sage: g.order() 24 sage: g.size() 36 sage: g.is_planar() True sage: g.is_regular(3) True sage: g.subgraph_search(graphs.CycleGraph(4)) is None # needs sage.modules True sage: g.subgraph_search(graphs.CycleGraph(8)) is None # needs sage.modules True sage: g.subgraph_search(graphs.CycleGraph(16)) # needs sage.modules Subgraph of (Markstroem Graph): Graph on 16 vertices
- static MathonPseudocyclicMergingGraph(M, t)#
Mathon’s merging of classes in a pseudo-cyclic 3-class association scheme
Construct strongly regular graphs from p.97 of [BL1984].
INPUT:
M– the list of matrices in a pseudo-cyclic 3-class association scheme. The identity matrix must be the first entry.t(integer) – the number of the graph, from 0 to 2.
See also
- static MathonPseudocyclicStronglyRegularGraph(t, G=None, L=None)#
Return a strongly regular graph on \((4t+1)(4t-1)^2\) vertices from [Mat1978].
Let \(4t-1\) be a prime power, and \(4t+1\) be such that there exists a strongly regular graph \(G\) with parameters \((4t+1,2t,t-1,t)\). In particular, \(4t+1\) must be a sum of two squares [Mat1978]. With this input, Mathon [Mat1978] gives a construction of a strongly regular graph with parameters \((4 \mu + 1, 2 \mu, \mu-1, \mu)\), where \(\mu = t(4t(4t-1)-1)\). The construction is optionally parametrised by an a skew-symmetric Latin square of order \(4t+1\), with entries in \(-2t,...,-1,0,1,...,2t\).
Our implementation follows a description given in [ST1981].
INPUT:
t– a positive integerG– ifNone(default), try to construct the necessary graph with parameters \((4t+1,2t,t-1,t)\), otherwise use the user-supplied one, with vertices labelled from \(0\) to \(4t\).L– ifNone(default), construct a necessary skew Latin square, otherwise use the user-supplied one. Here non-isomorphic Latin squares – one constructed from \(Z/9Z\), and the other from \((Z/3Z)^2\) – lead to non-isomorphic graphs.
See also
EXAMPLES:
Using default
GandL.sage: from sage.graphs.generators.families import MathonPseudocyclicStronglyRegularGraph sage: G = MathonPseudocyclicStronglyRegularGraph(1); G # needs sage.modules sage.rings.finite_rings Mathon's PC SRG on 45 vertices: Graph on 45 vertices sage: G.is_strongly_regular(parameters=True) # needs sage.modules sage.rings.finite_rings (45, 22, 10, 11)
Supplying
GandL(constructed from the automorphism group ofG). The entries of L can’t be tested directly because there’s some unpredictability in the way that GAP chooses a representative inNormalSubgroups(), the function that underlies our ownnormal_subgroups()method:sage: # needs sage.groups sage.libs.gap sage.rings.finite_rings sage: G = graphs.PaleyGraph(9) sage: a = G.automorphism_group(partition=[sorted(G)]) sage: it = (x for x in a.normal_subgroups() if x.order() == 9) sage: subg = next(iter(it)) sage: r = [matrix(libgap.PermutationMat(libgap(z), 9).sage()) ....: for z in subg] sage: ff = list(map(lambda y: (y[0]-1,y[1]-1), ....: Permutation(map(lambda x: 1+r.index(x^-1), r)).cycle_tuples()[1:])) sage: L = sum(i*(r[a]-r[b]) for i,(a,b) in zip(range(1,len(ff)+1), ff)) sage: G.relabel(range(9)) sage: G3x3 = graphs.MathonPseudocyclicStronglyRegularGraph(2, G=G, L=L) sage: G3x3.is_strongly_regular(parameters=True) (441, 220, 109, 110) sage: G3x3.automorphism_group(algorithm="bliss").order() # optional - bliss 27 sage: G9 = graphs.MathonPseudocyclicStronglyRegularGraph(2) sage: G9.is_strongly_regular(parameters=True) (441, 220, 109, 110) sage: G9.automorphism_group(algorithm="bliss").order() # optional - bliss 9
- static MathonStronglyRegularGraph(t)#
Return one of Mathon’s graphs on 784 vertices.
INPUT:
t– integer; the number of the graph, from 0 to 2
EXAMPLES:
sage: # long time, needs sage.libs.gap sage: from sage.graphs.generators.smallgraphs import MathonStronglyRegularGraph sage: G = MathonStronglyRegularGraph(0) sage: G.is_strongly_regular(parameters=True) (784, 243, 82, 72)
- static McGeeGraph(embedding=2)#
Return the McGee Graph.
See the Wikipedia article McGee_graph.
INPUT:
embedding– integer (default:2); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.McGeeGraph() sage: g.order() 24 sage: g.size() 36 sage: g.girth() 7 sage: g.diameter() 4 sage: g.show() # needs sage.plot sage: graphs.McGeeGraph(embedding=1).show() # long time # needs sage.plot
- static McLaughlinGraph()#
Return the McLaughlin Graph.
The McLaughlin Graph is the unique strongly regular graph of parameters \((275, 112, 30, 56)\).
For more information on the McLaughlin Graph, see its web page on Andries Brouwer’s website which gives the definition that this method implements.
Note
To create this graph you must have the gap_packages spkg installed.
EXAMPLES:
sage: g = graphs.McLaughlinGraph() # optional - gap_package_design sage: g.is_strongly_regular(parameters=True) # optional - gap_package_design (275, 112, 30, 56) sage: set(g.spectrum()) == {112, 2, -28} # optional - gap_package_design True
- static MeredithGraph()#
Return the Meredith Graph.
The Meredith Graph is a 4-regular 4-connected non-hamiltonian graph. For more information on the Meredith Graph, see the Wikipedia article Meredith_graph.
EXAMPLES:
sage: g = graphs.MeredithGraph() sage: g.is_regular(4) True sage: g.order() 70 sage: g.size() 140 sage: g.radius() 7 sage: g.diameter() 8 sage: g.girth() 4 sage: g.chromatic_number() 3 sage: g.is_hamiltonian() # long time # needs sage.numerical.mip False
- static MoebiusKantorGraph()#
Return a Möbius-Kantor Graph.
A Möbius-Kantor graph is a cubic symmetric graph. (See also the Heawood graph). It has 16 nodes and 24 edges. It is nonplanar and Hamiltonian. It has diameter 4, girth 6, and chromatic number 2. It is identical to the Generalized Petersen graph, P[8, 3].
For more details, see Möbius-Kantor Graph - from Wolfram MathWorld.
PLOTTING: See the plotting section for the generalized Petersen graphs.
EXAMPLES:
sage: MK = graphs.MoebiusKantorGraph() sage: MK Moebius-Kantor Graph: Graph on 16 vertices sage: MK.graph6_string() 'OhCGKE?O@?ACAC@I?Q_AS' sage: (graphs.MoebiusKantorGraph()).show() # long time # needs sage.plot
- static MoserSpindle()#
Return the Moser spindle.
For more information, see the Wikipedia article Moser_spindle.
EXAMPLES:
The Moser spindle is a planar graph having 7 vertices and 11 edges:
sage: # needs sage.symbolic sage: G = graphs.MoserSpindle(); G Moser spindle: Graph on 7 vertices sage: G.is_planar() True sage: G.order() 7 sage: G.size() 11
It is a Hamiltonian graph with radius 2, diameter 2, and girth 3:
sage: # needs sage.symbolic sage: G.is_hamiltonian() # needs sage.numerical.mip True sage: G.radius() 2 sage: G.diameter() 2 sage: G.girth() 3
The Moser spindle can be drawn in the plane as a unit distance graph, has chromatic number 4, and its automorphism group is isomorphic to the dihedral group \(D_4\):
sage: # needs sage.symbolic sage: pos = G.get_pos() sage: all(sum((ui-vi)**2 for ui, vi in zip(pos[u], pos[v])) == 1 ....: for u, v in G.edge_iterator(labels=None)) True sage: G.chromatic_number() 4 sage: ag = G.automorphism_group() sage: ag.is_isomorphic(DihedralGroup(4)) True
- static MuzychukS6Graph(n, d, Phi='fixed', Sigma='fixed', verbose=False)#
Return a strongly regular graph of S6 type from [Muz2007] on \(n^d((n^d-1)/(n-1)+1)\) vertices.
The construction depends upon a number of parameters, two of them, \(n\) and \(d\), mandatory, and \(\Phi\) and \(\Sigma\) mappings defined in [Muz2007]. These graphs have parameters \((mn^d, n^{d-1}(m-1) - 1,\mu - 2,\mu)\), where \(\mu=\frac{n^{d-1}-1}{n-1}n^{d-1}\) and \(m:=\frac{n^d-1}{n-1}+1\).
Some details on \(\Phi\) and \(\Sigma\) are as follows. Let \(L\) be the complete graph on \(M:=\{0,..., m-1\}\) with the matching \(\{(2i,2i+1) | i=0,...,m/2\}\) removed. Then one arbitrarily chooses injections \(\Phi_i\) from the edges of \(L\) on \(i \in M\) into sets of parallel classes of affine \(d\)-dimensional designs; our implementation uses the designs of hyperplanes in \(d\)-dimensional affine geometries over \(GF(n)\). Finally, for each edge \(ij\) of \(L\) one arbitrarily chooses bijections \(\Sigma_{ij}\) between \(\Phi_i\) and \(\Phi_j\). More details, in particular how these choices lead to non-isomorphic graphs, are in [Muz2007].
INPUT:
n(integer) – a prime powerd(integer) – must be odd if \(n\) is oddPhiis an optional parameter of the construction; it must be either'fixed'– this will generate fixed default \(\Phi_i\), for \(i \in M\), or'random'– \(\Phi_i\) are generated at random, orA dictionary describing the functions \(\Phi_i\); for \(i \in M\), Phi[(i, T)] in \(M\), for each edge T of \(L\) on \(i\). Also, each \(\Phi_i\) must be injective.
Sigmais an optional parameter of the construction; it must be either'fixed'– this will generate a fixed default \(\Sigma\), or'random'– \(\Sigma\) is generated at random.
verbose(Boolean) – default isFalse. IfTrue, print progress information
See also
Todo
Implement the possibility to explicitly supply the parameter \(\Sigma\) of the construction.
EXAMPLES:
sage: # needs sage.combinat sage.modules sage.rings.finite_rings sage: graphs.MuzychukS6Graph(3, 3).is_strongly_regular(parameters=True) (378, 116, 34, 36) sage: phi = {(2,(0,2)):0, (1,(1,3)):1, (0,(0,3)):1, (2,(1,2)):1, ....: (1,(1,2)):0, (0,(0,2)):0, (3,(0,3)):0, (3,(1,3)):1} sage: graphs.MuzychukS6Graph(2, 2, Phi=phi).is_strongly_regular(parameters=True) (16, 5, 0, 2)
- static MycielskiGraph(k=1, relabel=True)#
Returns the \(k\)-th Mycielski Graph.
The graph \(M_k\) is triangle-free and has chromatic number equal to \(k\). These graphs show, constructively, that there are triangle-free graphs with arbitrarily high chromatic number.
The Mycielski graphs are built recursively starting with \(M_0\), an empty graph; \(M_1\), a single vertex graph; and \(M_2\) is the graph \(K_2\). \(M_{k+1}\) is then built from \(M_k\) as follows:
If the vertices of \(M_k\) are \(v_1,\ldots,v_n\), then the vertices of \(M_{k+1}\) are \(v_1,\ldots,v_n,w_1,\ldots,w_n,z\). Vertices \(v_1,\ldots,v_n\) induce a copy of \(M_k\). Vertices \(w_1,\ldots,w_n\) are an independent set. Vertex \(z\) is adjacent to all the \(w_i\)-vertices. Finally, vertex \(w_i\) is adjacent to vertex \(v_j\) iff \(v_i\) is adjacent to \(v_j\).
INPUT:
kNumber of steps in the construction process.relabelRelabel the vertices so their names are the integersrange(n)wherenis the number of vertices in the graph.
EXAMPLES:
The Mycielski graph \(M_k\) is triangle-free and has chromatic number equal to \(k\).
sage: g = graphs.MycielskiGraph(5) sage: g.is_triangle_free() True sage: g.chromatic_number() 5
The graphs \(M_4\) is (isomorphic to) the Grotzsch graph.
sage: g = graphs.MycielskiGraph(4) sage: g.is_isomorphic(graphs.GrotzschGraph()) True
REFERENCES:
[1] Weisstein, Eric W. “Mycielski Graph.” From MathWorld–A Wolfram Web Resource. http://mathworld.wolfram.com/MycielskiGraph.html
- static MycielskiStep(g)#
Perform one iteration of the Mycielski construction.
See the documentation for
MycielskiGraphwhich uses this method. We expose it to all users in case they may find it useful.EXAMPLE. One iteration of the Mycielski step applied to the 5-cycle yields a graph isomorphic to the Grotzsch graph
sage: g = graphs.CycleGraph(5) sage: h = graphs.MycielskiStep(g) sage: h.is_isomorphic(graphs.GrotzschGraph()) True
- static NKStarGraph(n, k)#
Returns the (n,k)-star graph.
The vertices of the (n,k)-star graph are the set of all arrangements of n symbols into labels of length k. There are two adjacency rules for the (n,k)-star graph. First, two vertices are adjacent if one can be obtained from the other by swapping the first symbol with another symbol. Second, two vertices are adjacent if one can be obtained from the other by swapping the first symbol with an external symbol (a symbol not used in the original label).
INPUT:
nk
EXAMPLES:
sage: g = graphs.NKStarGraph(4,2) sage: g.plot() # long time # needs sage.plot Graphics object consisting of 31 graphics primitives
REFERENCES:
Wei-Kuo, Chiang, and Chen Rong-Jaye. “The (n, k)-star graph: A generalized star graph.” Information Processing Letters 56, no. 5 (December 8, 1995): 259-264.
AUTHORS:
Michael Yurko (2009-09-01)
- static NStarGraph(n)#
Returns the n-star graph.
The vertices of the n-star graph are the set of permutations on n symbols. There is an edge between two vertices if their labels differ only in the first and one other position.
INPUT:
n
EXAMPLES:
sage: g = graphs.NStarGraph(4) sage: g.plot() # long time # needs sage.plot Graphics object consisting of 61 graphics primitives
REFERENCES:
S.B. Akers, D. Horel and B. Krishnamurthy, The star graph: An attractive alternative to the previous n-cube. In: Proc. Internat. Conf. on Parallel Processing (1987), pp. 393–400.
AUTHORS:
Michael Yurko (2009-09-01)
- static NauruGraph(embedding=2)#
Return the Nauru Graph.
See the Wikipedia article Nauru_graph.
INPUT:
embedding– integer (default:2); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: g = graphs.NauruGraph() sage: g.order() 24 sage: g.size() 36 sage: g.girth() 6 sage: g.diameter() 4 sage: g.show() # needs sage.plot sage: graphs.NauruGraph(embedding=1).show() # long time # needs sage.plot
- static NonisotropicOrthogonalPolarGraph(m, q, sign='+', perp=None)#
Return the Graph \(NO^{\epsilon,\perp}_{m}(q)\)
Let the vectorspace of dimension \(m\) over \(F_q\) be endowed with a nondegenerate quadratic form \(F\), of type
signfor \(m\) even.\(m\) even: assume further that \(q=2\) or \(3\). Returns the graph of the points (in the underlying projective space) \(x\) satisfying \(F(x)=1\), with adjacency given by orthogonality w.r.t. \(F\). Parameter
perpis ignored.\(m\) odd: if
perpis notNone, then we assume that \(q=5\) and return the graph of the points \(x\) satisfying \(F(x)=\pm 1\) ifsign="+", respectively \(F(x) \in \{2,3\}\) ifsign="-", with adjacency given by orthogonality w.r.t. \(F\) (cf. Sect 7.D of [BL1984]). Otherwise return the graph of nongenerate hyperplanes of typesign, adjacent whenever the intersection is degenerate (cf. Sect. 7.C of [BL1984]). Note that for \(q=2\) one will get a complete graph.
For more information, see Sect. 9.9 of [BH2012] and [BL1984]. Note that the page of Andries Brouwer’s website uses different notation.
INPUT:
m– integer; half the dimension of the underlying vectorspaceq– a power of a prime number, the size of the underlying fieldsign– string (default:"+"); must be either"+"or"-"
EXAMPLES:
\(NO^-(4,2)\) is isomorphic to Petersen graph:
sage: g = graphs.NonisotropicOrthogonalPolarGraph(4,2,'-'); g # needs sage.libs.gap NO^-(4, 2): Graph on 10 vertices sage: g.is_strongly_regular(parameters=True) # needs sage.libs.gap (10, 3, 0, 1)
\(NO^-(6,2)\) and \(NO^+(6,2)\):
sage: # needs sage.libs.gap sage: g = graphs.NonisotropicOrthogonalPolarGraph(6,2,'-') sage: g.is_strongly_regular(parameters=True) (36, 15, 6, 6) sage: g = graphs.NonisotropicOrthogonalPolarGraph(6,2,'+'); g NO^+(6, 2): Graph on 28 vertices sage: g.is_strongly_regular(parameters=True) (28, 15, 6, 10)
\(NO^+(8,2)\):
sage: g = graphs.NonisotropicOrthogonalPolarGraph(8,2,'+') # needs sage.libs.gap sage: g.is_strongly_regular(parameters=True) # needs sage.libs.gap (120, 63, 30, 36)
Wilbrink’s graphs for \(q=5\):
sage: # needs sage.libs.gap sage: g = graphs.NonisotropicOrthogonalPolarGraph(5,5,perp=1) sage: g.is_strongly_regular(parameters=True) # long time (325, 60, 15, 10) sage: g = graphs.NonisotropicOrthogonalPolarGraph(5,5,'-',perp=1) sage: g.is_strongly_regular(parameters=True) # long time (300, 65, 10, 15)
Wilbrink’s graphs:
sage: # needs sage.libs.gap sage: g = graphs.NonisotropicOrthogonalPolarGraph(5,4,'+') sage: g.is_strongly_regular(parameters=True) (136, 75, 42, 40) sage: g = graphs.NonisotropicOrthogonalPolarGraph(5,4,'-') sage: g.is_strongly_regular(parameters=True) (120, 51, 18, 24) sage: g = graphs.NonisotropicOrthogonalPolarGraph(7,4,'+'); g # not tested (long time) NO^+(7, 4): Graph on 2080 vertices sage: g.is_strongly_regular(parameters=True) # not tested (long time) (2080, 1071, 558, 544)
- static NonisotropicUnitaryPolarGraph(m, q)#
Return the Graph \(NU(m,q)\).
This returns the graph on nonisotropic, with respect to a nondegenerate Hermitean form, points of the \((m-1)\)-dimensional projective space over \(F_q\), with points adjacent whenever they lie on a tangent (to the set of isotropic points) line. For more information, see Sect. 9.9 of [BH2012] and series C14 in [Hub1975].
INPUT:
m,q– integers; \(q\) must be a prime power
EXAMPLES:
sage: g = graphs.NonisotropicUnitaryPolarGraph(5,2); g # needs sage.libs.gap NU(5, 2): Graph on 176 vertices sage: g.is_strongly_regular(parameters=True) # needs sage.libs.gap (176, 135, 102, 108)
- static Nowhere0WordsTwoWeightCodeGraph(q, hyperoval=None, field=None, check_hyperoval=True)#
Return the subgraph of nowhere 0 words from two-weight code of projective plane hyperoval.
Let \(q=2^k\) and \(\Pi=PG(2,q)\). Fix a hyperoval \(O \subset \Pi\). Let \(V=F_q^3\) and \(C\) the two-weight 3-dimensional linear code over \(F_q\) with words \(c(v)\) obtained from \(v\in V\) by computing
\[c(v)=(\langle v,o_1 \rangle,...,\langle v,o_{q+2} \rangle), o_j \in O.\]\(C\) contains \(q(q-1)^2/2\) words without 0 entries. The subgraph of the strongly regular graph of \(C\) induced on the latter words is also strongly regular, assuming \(q>4\). This is a construction due to A.E.Brouwer [Bro2016], and leads to graphs with parameters also given by a construction in [HHL2009]. According to [Bro2016], these two constructions are likely to produce isomorphic graphs.
INPUT:
q– a power of twohyperoval– a hyperoval (i.e. a complete 2-arc; a set of points in the plane meeting every line in 0 or 2 points) in \(PG(2,q)\) over the fieldfield. Each point ofhyperovalmust be a length 3 vector overfieldwith 1st non-0 coordinate equal to 1. By default,hyperovalandfieldare not specified, and constructed on the fly. In particular,hyperovalwe build is the classical one, i.e. a conic with the point of intersection of its tangent lines.field– an instance of a finite field of order \(q\), must be provided ifhyperovalis provided.check_hyperoval– boolean (default:True); whether to checkhyperovalfor correctness or not
See also
EXAMPLES:
using the built-in construction:
sage: # needs sage.combinat sage.rings.finite_rings sage: g = graphs.Nowhere0WordsTwoWeightCodeGraph(8); g Nowhere0WordsTwoWeightCodeGraph(8): Graph on 196 vertices sage: g.is_strongly_regular(parameters=True) (196, 60, 14, 20) sage: g = graphs.Nowhere0WordsTwoWeightCodeGraph(16) # not tested (long time) sage: g.is_strongly_regular(parameters=True) # not tested (long time) (1800, 728, 268, 312)
supplying your own hyperoval:
sage: # needs sage.combinat sage.rings.finite_rings sage: F = GF(8) sage: O = [vector(F,(0,0,1)),vector(F,(0,1,0))] + [vector(F, (1,x^2,x)) ....: for x in F] sage: g = graphs.Nowhere0WordsTwoWeightCodeGraph(8,hyperoval=O,field=F); g Nowhere0WordsTwoWeightCodeGraph(8): Graph on 196 vertices sage: g.is_strongly_regular(parameters=True) (196, 60, 14, 20)
- static OctahedralGraph()#
Return an Octahedral graph (with 6 nodes).
The regular octahedron is an 8-sided polyhedron with triangular faces. The octahedral graph corresponds to the connectivity of the vertices of the octahedron. It is the line graph of the tetrahedral graph. The octahedral is symmetric, so the spring-layout algorithm will be very effective for display.
PLOTTING: The Octahedral graph should be viewed in 3 dimensions. We choose to use a planar embedding of the graph. We hope to add rotatable, 3-dimensional viewing in the future. In such a case, a argument will be added to select the desired layout.
EXAMPLES:
Construct and show an Octahedral graph:
sage: g = graphs.OctahedralGraph() sage: g.show() # long time # needs sage.plot
Create several octahedral graphs in a Sage graphics array They will be drawn differently due to the use of the spring-layout algorithm:
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.OctahedralGraph() ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static OddGraph(n)#
Returns the Odd Graph with parameter \(n\).
The Odd Graph with parameter \(n\) is defined as the Kneser Graph with parameters \(2n-1,n-1\). Equivalently, the Odd Graph is the graph whose vertices are the \(n-1\)-subsets of \([0,1,\dots,2(n-1)]\), and such that two vertices are adjacent if their corresponding sets are disjoint.
For example, the Petersen Graph can be defined as the Odd Graph with parameter \(3\).
EXAMPLES:
sage: OG = graphs.OddGraph(3) sage: sorted(OG.vertex_iterator(), key=str) [{1, 2}, {1, 3}, {1, 4}, {1, 5}, {2, 3}, {2, 4}, {2, 5}, {3, 4}, {3, 5}, {4, 5}] sage: P = graphs.PetersenGraph() sage: P.is_isomorphic(OG) True
- static OrthogonalArrayBlockGraph(k, n, OA=None)#
Return the graph of an \(OA(k,n)\).
The intersection graph of the blocks of a transversal design with parameters \((k,n)\), or \(TD(k,n)\) for short, is a strongly regular graph (unless it is a complete graph). Its parameters \((v,k',\lambda,\mu)\) are determined by the parameters \(k,n\) via:
\[v=n^2, k'=k(n-1), \lambda=(k-1)(k-2)+n-2, \mu=k(k-1)\]As transversal designs and orthogonal arrays (OA for short) are equivalent objects, this graph can also be built from the blocks of an \(OA(k,n)\), two of them being adjacent if one of their coordinates match.
For more information on these graphs, see Andries Brouwer’s page on Orthogonal Array graphs.
Warning
Brouwer’s website uses the notation \(OA(n,k)\) instead of \(OA(k,n)\)
For given parameters \(k\) and \(n\) there can be many \(OA(k,n)\) : the graphs returned are not uniquely defined by their parameters (see the examples below).
If the function is called only with the parameter
kandnthe results might be different with two versions of Sage, or even worse : some could not be available anymore.
INPUT:
k,n(integers)OA– An orthogonal array. If set toNone(default) thenorthogonal_array()is called to compute an \(OA(k,n)\).
EXAMPLES:
sage: # needs sage.modules sage: G = graphs.OrthogonalArrayBlockGraph(5,5); G # needs sage.schemes OA(5,5): Graph on 25 vertices sage: G.is_strongly_regular(parameters=True) # needs sage.schemes (25, 20, 15, 20) sage: G = graphs.OrthogonalArrayBlockGraph(4,10); G OA(4,10): Graph on 100 vertices sage: G.is_strongly_regular(parameters=True) (100, 36, 14, 12)
Two graphs built from different orthogonal arrays are also different:
sage: # needs sage.modules sage: k = 4; n = 10 sage: OAa = designs.orthogonal_arrays.build(k,n) sage: OAb = [[(x+1)%n for x in R] for R in OAa] sage: set(map(tuple,OAa)) == set(map(tuple,OAb)) False sage: Ga = graphs.OrthogonalArrayBlockGraph(k, n, OAa) sage: Gb = graphs.OrthogonalArrayBlockGraph(k, n, OAb) sage: Ga == Gb False
As
OAbwas obtained fromOAaby a relabelling the two graphs are isomorphic:sage: Ga.is_isomorphic(Gb) # needs sage.modules True
But there are examples of \(OA(k,n)\) for which the resulting graphs are not isomorphic:
sage: oa0 = [[0, 0, 1], [0, 1, 3], [0, 2, 0], [0, 3, 2], ....: [1, 0, 3], [1, 1, 1], [1, 2, 2], [1, 3, 0], ....: [2, 0, 0], [2, 1, 2], [2, 2, 1], [2, 3, 3], ....: [3, 0, 2], [3, 1, 0], [3, 2, 3], [3, 3, 1]] sage: oa1 = [[0, 0, 1], [0, 1, 0], [0, 2, 3], [0, 3, 2], ....: [1, 0, 3], [1, 1, 2], [1, 2, 0], [1, 3, 1], ....: [2, 0, 0], [2, 1, 1], [2, 2, 2], [2, 3, 3], ....: [3, 0, 2], [3, 1, 3], [3, 2, 1], [3, 3, 0]] sage: g0 = graphs.OrthogonalArrayBlockGraph(3, 4, oa0) # needs sage.modules sage: g1 = graphs.OrthogonalArrayBlockGraph(3, 4, oa1) # needs sage.modules sage: g0.is_isomorphic(g1) # needs sage.modules False
But nevertheless isospectral:
sage: g0.spectrum() # needs sage.modules sage.rings.number_field [9, 1, 1, 1, 1, 1, 1, 1, 1, 1, -3, -3, -3, -3, -3, -3] sage: g1.spectrum() # needs sage.modules sage.rings.number_field [9, 1, 1, 1, 1, 1, 1, 1, 1, 1, -3, -3, -3, -3, -3, -3]
Note that the graph
g0is actually isomorphic to the affine polar graph \(VO^+(4,2)\):sage: graphs.AffineOrthogonalPolarGraph(4,2,'+').is_isomorphic(g0) # needs sage.libs.gap sage.modules True
- static OrthogonalDualPolarGraph(e, d, q)#
Return the dual polar graph on \(GO^e(n,q)\) of diameter \(d\).
The value of \(n\) is determined by \(d\) and \(e\).
The graph is distance-regular with classical parameters \((d, q, 0, q^e)\).
INPUT:
e– integer; type of the orthogonal polar space to consider; must be \(-1, 0\) or \(1\).d– integer; diameter of the graphq– integer; prime power; order of the finite field over which to build the polar space
EXAMPLES:
sage: # needs sage.libs.gap sage: G = graphs.OrthogonalDualPolarGraph(1,3,2) sage: G.is_distance_regular(True) ([7, 6, 4, None], [None, 1, 3, 7]) sage: G = graphs.OrthogonalDualPolarGraph(0,3,3) # long time sage: G.is_distance_regular(True) # long time ([39, 36, 27, None], [None, 1, 4, 13]) sage: G.order() # long time 1120
REFERENCES:
See [BCN1989] pp. 274-279 or [VDKT2016] p. 22.
- static OrthogonalPolarGraph(m, q, sign='+')#
Return the Orthogonal Polar Graph \(O^{\epsilon}(m,q)\).
For more information on Orthogonal Polar graphs, see the page of Andries Brouwer’s website.
INPUT:
m,q– integers; \(q\) must be a prime powersign– string (default:"+"); must be"+"or"-"if \(m\) is even,"+"(default) otherwise
EXAMPLES:
sage: # needs sage.libs.gap sage: G = graphs.OrthogonalPolarGraph(6,3,"+"); G Orthogonal Polar Graph O^+(6, 3): Graph on 130 vertices sage: G.is_strongly_regular(parameters=True) (130, 48, 20, 16) sage: G = graphs.OrthogonalPolarGraph(6,3,"-"); G Orthogonal Polar Graph O^-(6, 3): Graph on 112 vertices sage: G.is_strongly_regular(parameters=True) (112, 30, 2, 10) sage: G = graphs.OrthogonalPolarGraph(5,3); G Orthogonal Polar Graph O(5, 3): Graph on 40 vertices sage: G.is_strongly_regular(parameters=True) (40, 12, 2, 4) sage: G = graphs.OrthogonalPolarGraph(8,2,"+"); G Orthogonal Polar Graph O^+(8, 2): Graph on 135 vertices sage: G.is_strongly_regular(parameters=True) (135, 70, 37, 35) sage: G = graphs.OrthogonalPolarGraph(8,2,"-"); G Orthogonal Polar Graph O^-(8, 2): Graph on 119 vertices sage: G.is_strongly_regular(parameters=True) (119, 54, 21, 27)
- static PaleyGraph(q)#
Paley graph with \(q\) vertices
Parameter \(q\) must be the power of a prime number and congruent to 1 mod 4.
EXAMPLES:
sage: G = graphs.PaleyGraph(9); G # needs sage.rings.finite_rings Paley graph with parameter 9: Graph on 9 vertices sage: G.is_regular() # needs sage.rings.finite_rings True
A Paley graph is always self-complementary:
sage: G.is_self_complementary() # needs sage.rings.finite_rings True
- static PappusGraph()#
Return the Pappus graph, a graph on 18 vertices.
The Pappus graph is cubic, symmetric, and distance-regular.
EXAMPLES:
sage: G = graphs.PappusGraph() sage: G.show() # long time # needs sage.plot sage: L = graphs.LCFGraph(18, [5,7,-7,7,-7,-5], 3) # needs networkx sage: L.show() # long time # needs networkx sage.plot sage: G.is_isomorphic(L) # needs networkx True
- static PasechnikGraph(n)#
Pasechnik strongly regular graph on \((4n-1)^2\) vertices
A strongly regular graph with parameters of the orthogonal array graph
OrthogonalArrayBlockGraph(), also known as pseudo Latin squares graph \(L_{2n-1}(4n-1)\), constructed from a skew Hadamard matrix of order \(4n\) following [Pas1992].See also
EXAMPLES:
sage: graphs.PasechnikGraph(4).is_strongly_regular(parameters=True) # needs sage.combinat sage.modules (225, 98, 43, 42) sage: graphs.PasechnikGraph(5).is_strongly_regular(parameters=True) # long time, needs sage.combinat sage.modules (361, 162, 73, 72) sage: graphs.PasechnikGraph(9).is_strongly_regular(parameters=True) # not tested (1225, 578, 273, 272)
- static PathGraph(n, pos=None)#
Return a path graph with \(n\) nodes.
A path graph is a graph where all inner nodes are connected to their two neighbors and the two end-nodes are connected to their one inner neighbors (i.e.: a cycle graph without the first and last node connected).
INPUT:
n– number of nodes of the path graphpos– string (default:None); indicates the embedding to use between ‘circle’, ‘line’ or the default algorithm. See the plotting section below for more detail.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the graph may be drawn in one of two ways: The ‘line’ argument will draw the graph in a horizontal line (left to right) if there are less than 11 nodes. Otherwise the ‘line’ argument will append horizontal lines of length 10 nodes below, alternating left to right and right to left. The ‘circle’ argument will cause the graph to be drawn in a cycle-shape, with the first node at the top and then about the circle in a clockwise manner. By default (without an appropriate string argument) the graph will be drawn as a ‘circle’ if \(10 < n < 41\) and as a ‘line’ for all other \(n\).
EXAMPLES: Show default drawing by size: ‘line’: \(n \leq 10\)
sage: p = graphs.PathGraph(10) sage: p.show() # long time # needs sage.plot
‘circle’: \(10 < n < 41\)
sage: q = graphs.PathGraph(25) sage: q.show() # long time # needs sage.plot
‘line’: \(n \geq 41\)
sage: r = graphs.PathGraph(55) sage: r.show() # long time # needs sage.plot
Override the default drawing:
sage: s = graphs.PathGraph(5,'circle') sage: s.show() # long time # needs sage.plot
- static PerkelGraph()#
Return the Perkel Graph.
The Perkel Graph is a 6-regular graph with \(57\) vertices and \(171\) edges. It is the unique distance-regular graph with intersection array \((6,5,2;1,1,3)\). For more information, see the Wikipedia article Perkel_graph or https://www.win.tue.nl/~aeb/graphs/Perkel.html.
EXAMPLES:
sage: g = graphs.PerkelGraph(); g Perkel Graph: Graph on 57 vertices sage: g.is_distance_regular(parameters=True) ([6, 5, 2, None], [None, 1, 1, 3])
- static PermutationGraph(second_permutation, first_permutation=None)#
Build a permutation graph from one permutation or from two lists.
Definition:
If \(\sigma\) is a permutation of \(\{ 1, 2, \ldots, n \}\), then the permutation graph of \(\sigma\) is the graph on vertex set \(\{ 1, 2, \ldots, n \}\) in which two vertices \(i\) and \(j\) satisfying \(i < j\) are connected by an edge if and only if \(\sigma^{-1}(i) > \sigma^{-1}(j)\). A visual way to construct this graph is as follows:
Take two horizontal lines in the euclidean plane, and mark points \(1, ..., n\) from left to right on the first of them. On the second one, still from left to right, mark \(n\) points \(\sigma(1), \sigma(2), \ldots, \sigma(n)\). Now, link by a segment the two points marked with \(1\), then link together the points marked with \(2\), and so on. The permutation graph of \(\sigma\) is the intersection graph of those segments: there exists a vertex in this graph for each element from \(1\) to \(n\), two vertices \(i, j\) being adjacent if the segments \(i\) and \(j\) cross each other.
The set of edges of the permutation graph can thus be identified with the set of inversions of the inverse of the given permutation \(\sigma\).
A more general notion of permutation graph can be defined as follows: If \(S\) is a set, and \((a_1, a_2, \ldots, a_n)\) and \((b_1, b_2, \ldots, b_n)\) are two lists of elements of \(S\), each of which lists contains every element of \(S\) exactly once, then the permutation graph defined by these two lists is the graph on the vertex set \(S\) in which two vertices \(i\) and \(j\) are connected by an edge if and only if the order in which these vertices appear in the list \((a_1, a_2, \ldots, a_n)\) is the opposite of the order in which they appear in the list \((b_1, b_2, \ldots, b_n)\). When \((a_1, a_2, \ldots, a_n) = (1, 2, \ldots, n)\), this graph is the permutation graph of the permutation \((b_1, b_2, \ldots, b_n) \in S_n\). Notice that \(S\) does not have to be a set of integers here, but can be a set of strings, tuples, or anything else. We can still use the above visual description to construct the permutation graph, but now we have to mark points \(a_1, a_2, \ldots, a_n\) from left to right on the first horizontal line and points \(b_1, b_2, \ldots, b_n\) from left to right on the second horizontal line.
INPUT:
second_permutation– the unique permutation/list defining the graph, or the second of the two (if the graph is to be built from two permutations/lists).first_permutation(optional) – the first of the two permutations/lists from which the graph should be built, if it is to be built from two permutations/lists.When
first_permutation is None(default), it is set to be equal tosorted(second_permutation), which yields the expected ordering when the elements of the graph are integers.
See also
Recognition of Permutation graphs in the
comparability module.Drawings of permutation graphs as intersection graphs of segments is possible through the
show()method ofPermutationobjects.The correct argument to use in this case is
show(representation = "braid").
EXAMPLES:
sage: p = Permutations(5).random_element() sage: PG = graphs.PermutationGraph(p) sage: edges = PG.edges(sort=True, labels=False) sage: set(edges) == set(p.inverse().inversions()) True sage: PG = graphs.PermutationGraph([3,4,5,1,2]) sage: sorted(PG.edges(sort=True)) [(1, 3, None), (1, 4, None), (1, 5, None), (2, 3, None), (2, 4, None), (2, 5, None)] sage: PG = graphs.PermutationGraph([3,4,5,1,2], [1,4,2,5,3]) sage: sorted(PG.edges(sort=True)) [(1, 3, None), (1, 4, None), (1, 5, None), (2, 3, None), (2, 5, None), (3, 4, None), (3, 5, None)] sage: PG = graphs.PermutationGraph([1,4,2,5,3], [3,4,5,1,2]) sage: sorted(PG.edges(sort=True)) [(1, 3, None), (1, 4, None), (1, 5, None), (2, 3, None), (2, 5, None), (3, 4, None), (3, 5, None)] sage: PG = graphs.PermutationGraph(Permutation([1,3,2]), Permutation([1,2,3])) sage: sorted(PG.edges(sort=True)) [(2, 3, None)] sage: graphs.PermutationGraph([]).edges(sort=True) [] sage: graphs.PermutationGraph([], []).edges(sort=True) [] sage: PG = graphs.PermutationGraph("graph", "phrag") sage: sorted(PG.edges(sort=True)) [('a', 'g', None), ('a', 'h', None), ('a', 'p', None), ('g', 'h', None), ('g', 'p', None), ('g', 'r', None), ('h', 'r', None), ('p', 'r', None)]
- static PetersenGraph()#
Return the Petersen Graph.
The Petersen Graph is a named graph that consists of 10 vertices and 15 edges, usually drawn as a five-point star embedded in a pentagon.
The Petersen Graph is a common counterexample. For example, it is not Hamiltonian.
PLOTTING: See the plotting section for the generalized Petersen graphs.
EXAMPLES: We compare below the Petersen graph with the default spring-layout versus a planned position dictionary of \((x, y)\) tuples:
sage: petersen_spring = Graph({0:[1,4,5], 1:[0,2,6], 2:[1,3,7], ....: 3:[2,4,8], 4:[0,3,9], 5:[0,7,8], ....: 6:[1,8,9], 7:[2,5,9], 8:[3,5,6], ....: 9:[4,6,7]}) sage: petersen_spring.show() # long time # needs sage.plot sage: petersen_database = graphs.PetersenGraph() sage: petersen_database.show() # long time # needs sage.plot
- static PoussinGraph()#
Return the Poussin Graph.
For more information on the Poussin Graph, see its corresponding Wolfram page.
EXAMPLES:
sage: g = graphs.PoussinGraph() sage: g.order() 15 sage: g.is_planar() True
- static QueenGraph(dim_list, radius=None, relabel=False)#
Return the \(d\)-dimensional Queen Graph with prescribed dimensions.
The 2-dimensional Queen Graph of parameters \(n\) and \(m\) is a graph with \(nm\) vertices in which each vertex represents a square in an \(n \times m\) chessboard, and each edge corresponds to a legal move by a queen.
The \(d\)-dimensional Queen Graph with \(d >= 2\) has for vertex set the cells of a \(d\)-dimensional grid with prescribed dimensions, and each edge corresponds to a legal move by a queen in either one or two dimensions.
All 2-dimensional Queen Graphs are Hamiltonian and biconnected. The chromatic number of a \((n,n)\)-Queen Graph is at least \(n\), and it is exactly \(n\) when \(n\equiv 1,5 \bmod{6}\).
INPUT:
dim_list– iterable (list, set, dict); provides the dimensions \(n_1, n_2, \ldots, n_d\), with \(n_i \geq 1\), of the chessboardradius– integer (default:None); by setting the radius to a positive integer, one may reduce the visibility of the queen to at mostradiussteps. When radius is 1, the resulting graph is a King Graph.relabel– boolean (default:False); indicates whether the vertices must be relabeled as integers
EXAMPLES:
The \((2,2)\)-Queen Graph is isomorphic to the complete graph on 4 vertices:
sage: G = graphs.QueenGraph([2, 2]) sage: G.is_isomorphic(graphs.CompleteGraph(4)) True
The Queen Graph with radius 1 is isomorphic to the King Graph:
sage: G = graphs.QueenGraph([4, 5], radius=1) sage: H = graphs.KingGraph([5, 4]) sage: G.is_isomorphic(H) True
Also True in higher dimensions:
sage: G = graphs.QueenGraph([3, 4, 5], radius=1) sage: H = graphs.KingGraph([5, 3, 4]) sage: G.is_isomorphic(H) True
The Queen Graph can be obtained from the Rook Graph and the Bishop Graph:
sage: for d in range(3,12): # long time ....: for r in range(1,d+1): ....: G = graphs.QueenGraph([d,d],radius=r) ....: H = graphs.RookGraph([d,d],radius=r) ....: B = graphs.BishopGraph([d,d],radius=r) ....: H.add_edges(B.edges(sort=False)) ....: if not G.is_isomorphic(H): ....: print("that's not good!")
- static RandomBarabasiAlbert(n, m, seed=None)#
Return a random graph created using the Barabasi-Albert preferential attachment model.
A graph with \(m\) vertices and no edges is initialized, and a graph of \(n\) vertices is grown by attaching new vertices each with \(m\) edges that are attached to existing vertices, preferentially with high degree.
INPUT:
n– number of vertices in the graphm– number of edges to attach from each new nodeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
We show the edge list of a random graph on 6 nodes with \(m = 2\):
sage: G = graphs.RandomBarabasiAlbert(6,2) # needs networkx sage: G.order(), G.size() # needs networkx (6, 8) sage: G.degree_sequence() # random # needs networkx [4, 3, 3, 2, 2, 2]
We plot a random graph on 12 nodes with \(m = 3\):
sage: ba = graphs.RandomBarabasiAlbert(12,3) # needs networkx sage: ba.show() # long time # needs networkx sage.plot
We view many random graphs using a graphics array:
sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(1,10): ....: k = graphs.RandomBarabasiAlbert(i+3, 3) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
When \(m = 1\), the generated graph is a tree:
sage: graphs.RandomBarabasiAlbert(6, 1).is_tree() # needs networkx True
- static RandomBicubicPlanar(n, seed=None)#
Return the graph of a random bipartite cubic map with \(3 n\) edges.
INPUT:
\(n\) – an integer (at least \(1\))
seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
OUTPUT:
a graph with multiple edges (no embedding is provided)
The algorithm used is described in [Sch1999]. This samples a random rooted bipartite cubic map, chosen uniformly at random.
First one creates a random binary tree with \(n\) vertices. Next one turns this into a blossoming tree (at random) and reads the contour word of this blossoming tree.
Then one performs a rotation on this word so that this becomes a balanced word. There are three ways to do that, one is picked at random. Then a graph is build from the balanced word by iterated closure (adding edges).
In the returned graph, the three edges incident to any given vertex are colored by the integers 0, 1 and 2.
See also
the auxiliary method
blossoming_contour()EXAMPLES:
sage: # needs sage.combinat sage: n = randint(200, 300) sage: G = graphs.RandomBicubicPlanar(n) sage: G.order() == 2*n True sage: G.size() == 3*n True sage: G.is_bipartite() and G.is_planar() and G.is_regular(3) True sage: dic = {'red': [v for v in G.vertices(sort=False) if v[0] == 'n'], ....: 'blue': [v for v in G.vertices(sort=False) if v[0] != 'n']} sage: G.plot(vertex_labels=False, vertex_size=20, vertex_colors=dic) # needs sage.plot Graphics object consisting of ... graphics primitives
- static RandomBipartite(n1, n2, p, set_position=False, seed=None)#
Returns a bipartite graph with \(n1+n2\) vertices such that any edge from \([n1]\) to \([n2]\) exists with probability \(p\).
INPUT:
n1, n2– Cardinalities of the two setsp– Probability for an edge to existset_position– boolean (defaultFalse); if set toTrue, we assign positions to the vertices so that the set of cardinality \(n1\) is on the line \(y=1\) and the set of cardinality \(n2\) is on the line \(y=0\).seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: g = graphs.RandomBipartite(5, 2, 0.5) # needs numpy sage: g.vertices(sort=True) # needs numpy [(0, 0), (0, 1), (0, 2), (0, 3), (0, 4), (1, 0), (1, 1)]
- static RandomBlockGraph(m, k, kmax=None, incidence_structure=False, seed=None)#
Return a Random Block Graph.
A block graph is a connected graph in which every biconnected component (block) is a clique.
See also
Wikipedia article Block_graph for more details on these graphs
is_block_graph()– test if a graph is a block graph
INPUT:
m– integer; number of blocks (at least one).k– integer; minimum number of vertices of a block (at least two).kmax– integer (default:None) By default, each block has \(k\) vertices. When the parameter \(kmax\) is specified (with \(kmax \geq k\)), the number of vertices of each block is randomly chosen between \(k\) and \(kmax\).incidence_structure– boolean (default:False) when set toTrue, the incidence structure of the graphs is returned instead of the graph itself, that is the list of the lists of vertices in each block. This is useful for the creation of some hypergraphs.seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
OUTPUT:
A Graph when
incidence_structure==False(default), and otherwise an incidence structure.EXAMPLES:
A block graph with a single block is a clique:
sage: B = graphs.RandomBlockGraph(1, 4) sage: B.is_clique() True
A block graph with blocks of order 2 is a tree:
sage: B = graphs.RandomBlockGraph(10, 2) sage: B.is_tree() True
Every biconnected component of a block graph is a clique:
sage: B = graphs.RandomBlockGraph(5, 3, kmax=6) sage: blocks,cuts = B.blocks_and_cut_vertices() sage: all(B.is_clique(block) for block in blocks) True
A block graph with blocks of order \(k\) has \(m*(k-1)+1\) vertices:
sage: m, k = 6, 4 sage: B = graphs.RandomBlockGraph(m, k) sage: B.order() == m*(k-1)+1 True
Test recognition methods:
sage: B = graphs.RandomBlockGraph(6, 2, kmax=6) sage: B.is_block_graph() True sage: B in graph_classes.Block True
Asking for the incidence structure:
sage: m, k = 6, 4 sage: IS = graphs.RandomBlockGraph(m, k, incidence_structure=True) sage: from sage.combinat.designs.incidence_structures import IncidenceStructure sage: IncidenceStructure(IS) # needs sage.modules Incidence structure with 19 points and 6 blocks sage: m*(k-1)+1 19
- static RandomBoundedToleranceGraph(n, seed=None)#
Return a random bounded tolerance graph.
The random tolerance graph is built from a random bounded tolerance representation by using the function \(ToleranceGraph\). This representation is a list \(((l_0,r_0,t_0), (l_1,r_1,t_1), ..., (l_k,r_k,t_k))\) where \(k = n-1\) and \(I_i = (l_i,r_i)\) denotes a random interval and \(t_i\) a random positive value less than or equal to the length of the interval \(I_i\). The width of the representation is limited to \(n^2 * 2^n\).
Note
The tolerance representation used to create the graph can be recovered using
get_vertex()orget_vertices().INPUT:
n– number of vertices of the random graph.seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
Every (bounded) tolerance graph is perfect. Hence, the chromatic number is equal to the clique number
sage: g = graphs.RandomBoundedToleranceGraph(8) sage: g.clique_number() == g.chromatic_number() True
- static RandomChordalGraph(n, algorithm='growing', k=None, l=None, f=None, s=None, seed=None)#
Return a random chordal graph of order
n.A Graph \(G\) is said to be chordal if it contains no induced hole (a cycle of length at least 4). Equivalently, \(G\) is chordal if it has a perfect elimination orderings, if each minimal separator is a clique, or if it is the intersection graphs of subtrees of a tree. See the Wikipedia article Chordal_graph.
This generator implements the algorithms proposed in [SHET2018] for generating random chordal graphs as the intersection graph of \(n\) subtrees of a tree of order \(n\).
The returned graph is not necessarily connected.
INPUT:
n– integer; the number of nodes of the graphalgorithm– string (default:"growing"); the choice of the algorithm for randomly selecting \(n\) subtrees of a random tree of order \(n\). Possible choices are:"growing"– for each subtree \(T_i\), the algorithm picks a size \(k_i\) randomly from \([1,k]\). Then a random node of \(T\) is chosen as the first node of \(T_i\). In each of the subsequent \(k_i - 1\) iterations, it picks a random node in the neighborhood of \(T_i\) and adds it to \(T_i\)."connecting"– for each subtree \(T_i\), it first selects \(k_i\) nodes of \(T\), where \(k_i\) is a random integer from a Poisson distribution with mean \(l\). \(T_i\) is then generated to be the minimal subtree containing the selected \(k_i\) nodes. This implies that a subtree will most likely have many more nodes than those selected initially, and this must be taken into consideration when choosing \(l\)."pruned"– for each subtree \(T_i\), it randomly selects a fraction \(f\) of the edges on the tree and removes them. The number of edges to delete, say \(l\), is calculated as \(\lfloor (n - 1) f \rfloor\), which will leave \(l + 1\) subtrees in total. Then, it determines the sizes of the \(l + 1\) subtrees and stores the distinct values. Finally, it picks a random size \(k_i\) from the set of largest \(100(1-s)\%\) of distinct values, and randomly chooses a subtree with size \(k_i\).
k– integer (default:None); maximum size of a subtree. If not specified (None), the maximum size is set to \(\sqrt{n}\). This parameter is used only whenalgorithm="growing". Seegrowing_subtrees()for more details.l– a strictly positive real number (default:None); mean of a Poisson distribution. If not specified, the mean in set to \(\log_2{n}\). This parameter is used only whenalgorithm="connecting". Seeconnecting_nodes()for more details.f– a rational number (default:None); the edge deletion fraction. This value must be chosen in \([0..1]\). If not specified, this parameter is set to \(\frac{1}{n-1}\). This parameter is used only whenalgorithm="pruned". Seepruned_tree()for more details.s– a real number between 0 and 1 (default:None); selection barrier for the size of trees. If not specified, this parameter is set to \(0.5\). This parameter is used only whenalgorithm="pruned". Seepruned_tree()for more details.seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: from sage.graphs.generators.random import RandomChordalGraph sage: T = RandomChordalGraph(20, algorithm="growing", k=5) sage: T.is_chordal() True sage: T = RandomChordalGraph(20, algorithm="connecting", l=3) # needs numpy sage: T.is_chordal() # needs numpy True sage: T = RandomChordalGraph(20, algorithm="pruned", f=1/3, s=.5) sage: T.is_chordal() True
See also
growing_subtrees()connecting_nodes()pruned_tree()
- static RandomGNM(n, m, dense=False, seed=None)#
Returns a graph randomly picked out of all graphs on n vertices with m edges.
INPUT:
n– number of vertices.m– number of edges.dense– whether to use NetworkX’sdense_gnm_random_graph()orgnm_random_graph()seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
We show the edge list of a random graph on 5 nodes with 10 edges:
sage: graphs.RandomGNM(5, 10).edges(sort=True, labels=False) # needs networkx [(0, 1), (0, 2), (0, 3), (0, 4), (1, 2), (1, 3), (1, 4), (2, 3), (2, 4), (3, 4)]
We plot a random graph on 12 nodes with m = 12:
sage: gnm = graphs.RandomGNM(12, 12) # needs networkx sage: gnm.show() # long time # needs networkx sage.plot
We view many random graphs using a graphics array:
sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.RandomGNM(i+3, i^2-i) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static RandomGNP(n, p, seed=None, fast=True, algorithm='Sage')#
Returns a random graph on \(n\) nodes. Each edge is inserted independently with probability \(p\).
INPUT:
n– number of nodes of the graphp– probability of an edgeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)fast– boolean set to True (default) to use the algorithm with time complexity in \(O(n+m)\) proposed in [BB2005a]. It is designed for generating large sparse graphs. It is faster than other algorithms for LARGE instances (try it to know whether it is useful for you).algorithm– By default (algorithm='Sage'), this function uses the algorithm implemented in`sage.graphs.graph_generators_pyx.pyx. Whenalgorithm='networkx', this function calls the NetworkX functionfast_gnp_random_graph, unlessfast=False, thengnp_random_graph. Try them to know which algorithm is the best for you. Thefastparameter is not taken into account by the ‘Sage’ algorithm so far.
REFERENCES:
PLOTTING: When plotting, this graph will use the default spring-layout algorithm, unless a position dictionary is specified.
EXAMPLES: We show the edge list of a random graph on 6 nodes with probability \(p = .4\):
sage: set_random_seed(0) sage: graphs.RandomGNP(6, .4).edges(sort=true, labels=False) [(0, 3), (1, 2), (2, 3), (2, 4)]
We plot a random graph on 12 nodes with probability \(p = .71\):
sage: gnp = graphs.RandomGNP(12,.71) sage: gnp.show() # long time # needs sage.plot
We view many random graphs using a graphics array:
sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.RandomGNP(i+3,.43) ....: g.append(k) sage: for i in range(3): # needs sage.plot ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) # needs sage.plot sage: G.show() # long time # needs sage.plot sage: graphs.RandomGNP(4,1) Complete graph: Graph on 4 vertices
- static RandomHolmeKim(n, m, p, seed=None)#
Return a random graph generated by the Holme and Kim algorithm for graphs with power law degree distribution and approximate average clustering.
INPUT:
n– number of verticesm– number of random edges to add for each new nodep– probability of adding a triangle after adding a random edgeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
From the NetworkX documentation: the average clustering has a hard time getting above a certain cutoff that depends on \(m\). This cutoff is often quite low. Note that the transitivity (fraction of triangles to possible triangles) seems to go down with network size. It is essentially the Barabasi-Albert growth model with an extra step that each random edge is followed by a chance of making an edge to one of its neighbors too (and thus a triangle). This algorithm improves on B-A in the sense that it enables a higher average clustering to be attained if desired. It seems possible to have a disconnected graph with this algorithm since the initial \(m\) nodes may not be all linked to a new node on the first iteration like the BA model.
EXAMPLES:
sage: G = graphs.RandomHolmeKim(12, 3, .3) # needs networkx sage: G.show() # long time # needs networkx sage.plot
REFERENCE:
- static RandomIntervalGraph(n, seed=None)#
Returns a random interval graph.
An interval graph is built from a list \((a_i,b_i)_{1\leq i \leq n}\) of intervals : to each interval of the list is associated one vertex, two vertices being adjacent if the two corresponding intervals intersect.
A random interval graph of order \(n\) is generated by picking random values for the \((a_i,b_j)\), each of the two coordinates being generated from the uniform distribution on the interval \([0,1]\).
This definitions follows [BF2001].
Note
The vertices are named 0, 1, 2, and so on. The intervals used to create the graph are saved with the graph and can be recovered using
get_vertex()orget_vertices().INPUT:
n– integer; the number of vertices in the random graphseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
As for any interval graph, the chromatic number is equal to the clique number
sage: g = graphs.RandomIntervalGraph(8) sage: g.clique_number() == g.chromatic_number() True
- static RandomKTree(n, k, seed=None)#
Return a random \(k\)-tree on \(n\) nodes numbered \(0\) through \(n-1\).
ALGORITHM:
The algorithm first generates a complete graph on \(k + 1\) vertices. Vertices are subsequently generated by randomly choosing one of the existing cliques in the graph, and creating a new clique by replacing one of the vertices in the selected clique with a newly created one.
INPUT:
n– number of vertices in the \(k\)-treek– within a clique each vertex is connected to \(k\) vertices. \(k\) also corresponds to the treewidth of the \(k\)-treeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: G = graphs.RandomKTree(50, 5) sage: G.treewidth() 5 sage: G.show() # not tested
- static RandomLobster(n, p, q, seed=None)#
Returns a random lobster.
A lobster is a tree that reduces to a caterpillar when pruning all leaf vertices. A caterpillar is a tree that reduces to a path when pruning all leaf vertices (q=0).
INPUT:
n- expected number of vertices in the backbonep- probability of adding an edge to the backboneq- probability of adding an edge (claw) to the armsseed- arandom.Randomseed or a Pythonintfor the random number generator (default:None).
EXAMPLES:
We check a random graph with 12 backbone nodes and probabilities \(p = 0.7\) and \(q = 0.3\):
sage: # needs networkx sage: G = graphs.RandomLobster(12, 0.7, 0.3) sage: leaves = [v for v in G.vertices(sort=False) if G.degree(v) == 1] sage: G.delete_vertices(leaves) # caterpillar sage: leaves = [v for v in G.vertices(sort=False) if G.degree(v) == 1] sage: G.delete_vertices(leaves) # path sage: s = G.degree_sequence() sage: if G: ....: if G.num_verts() == 1: ....: assert s == [0] ....: else: ....: assert s[-2:] == [1, 1] ....: assert all(d == 2 for d in s[:-2])
sage: G = graphs.RandomLobster(9, .6, .3) # needs networkx sage: G.show() # long time # needs networkx sage.plot
- static RandomNewmanWattsStrogatz(n, k, p, seed=None)#
Return a Newman-Watts-Strogatz small world random graph on \(n\) vertices.
From the NetworkX documentation: first create a ring over \(n\) nodes. Then each node in the ring is connected with its \(k\) nearest neighbors. Then shortcuts are created by adding new edges as follows: for each edge \(u-v\) in the underlying “\(n\)-ring with \(k\) nearest neighbors”; with probability \(p\) add a new edge \(u-w\) with randomly-chosen existing node \(w\). In contrast with
networkx.watts_strogatz_graph(), no edges are removed.INPUT:
n– number of verticesk– each vertex is connected to its \(k\) nearest neighborsp– the probability of adding a new edge for each edgeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
We check that the generated graph contains a cycle of order \(n\):
sage: # needs networkx sage: G = graphs.RandomNewmanWattsStrogatz(7, 2, 0.2) sage: G.order() 7 sage: C7 = graphs.CycleGraph(7) sage: G.subgraph_search(C7) Subgraph of (): Graph on 7 vertices sage: G.diameter() <= C7.diameter() True
sage: G = graphs.RandomNewmanWattsStrogatz(12, 2, .3) # needs networkx sage: G.show() # long time # needs networkx sage.plot
REFERENCE:
- static RandomPartialKTree(n, k, x, seed=None)#
Return a random partial \(k\)-tree on \(n\) nodes.
A partial \(k\)-tree is defined as a subgraph of a \(k\)-tree. This can also be described as a graph with treewidth at most \(k\).
INPUT:
n– number of vertices in the \(k\)-treek– within a clique each vertex is connected to \(k\) vertices. \(k\) also corresponds to the treewidth of the \(k\)-treex– how many edges are deleted from the \(k\)-treeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: G = graphs.RandomPartialKTree(50,5,2) sage: G.treewidth() 5 sage: G.show() # not tested
- static RandomRegular(d, n, seed=None)#
Return a random \(d\)-regular graph on \(n\) vertices, or
Falseon failure.Since every edge is incident to two vertices, \(n\times d\) must be even.
INPUT:
d– degreen– number of verticesseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
We check that a random graph with 8 nodes each of degree 3 is 3-regular:
sage: G = graphs.RandomRegular(3, 8) # needs networkx sage: G.is_regular(k=3) # needs networkx True sage: G.degree_histogram() # needs networkx [0, 0, 0, 8]
sage: G = graphs.RandomRegular(3, 20) # needs networkx sage: if G: # random output # long time, needs networkx sage.plot ....: G.show()
REFERENCES:
- static RandomRegularBipartite(n1, n2, d1, set_position=False, seed=None)#
Return a random regular bipartite graph on \(n1 + n2\) vertices.
The bipartite graph has \(n1 * d1\) edges. Hence, \(n2\) must divide \(n1 * d1\). Each vertex of the set of cardinality \(n1\) has degree \(d1\) (which can be at most \(n2\)) and each vertex in the set of cardinality \(n2\) has degree \((n1 * d1) / n2\). The bipartite graph has no multiple edges.
This generator implements an algorithm inspired by that of [MW1990] for the uniform generation of random regular bipartite graphs. It performs well when \(d1 = o(n2^{1/3})\) or (\(n2 - d1 = o(n2^{1/3})\)). In other cases, the running time can be huge. Note that the currently implemented algorithm does not generate uniformly random graphs.
INPUT:
n1, n2– number of vertices in each sided1– degree of the vertices in the set of cardinality \(n1\).set_position– boolean (defaultFalse); if set toTrue, we assign positions to the vertices so that the set of cardinality \(n1\) is on the line \(y=1\) and the set of cardinality \(n2\) is on the line \(y=0\).seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: g = graphs.RandomRegularBipartite(4, 6, 3) sage: g.order(), g.size() (10, 12) sage: set(g.degree()) {2, 3} sage: graphs.RandomRegularBipartite(1, 2, 2, set_position=True).get_pos() {0: (1, 1.0), 1: (0, 0), 2: (2.0, 0.0)} sage: graphs.RandomRegularBipartite(2, 1, 1, set_position=True).get_pos() {0: (0, 1), 1: (2.0, 1.0), 2: (1, 0.0)} sage: graphs.RandomRegularBipartite(2, 3, 3, set_position=True).get_pos() {0: (0, 1), 1: (3.0, 1.0), 2: (0, 0), 3: (1.5, 0.0), 4: (3.0, 0.0)} sage: graphs.RandomRegularBipartite(2, 3, 3, set_position=False).get_pos()
- static RandomShell(constructor, seed=None)#
Return a random shell graph for the constructor given.
INPUT:
constructor– a list of 3-tuples \((n, m, d)\), each representing a shell, where:n– the number of vertices in the shellm– the number of edges in the shelld– the ratio of inter (next) shell edges to intra shell edges
seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: G = graphs.RandomShell([(10,20,0.8),(20,40,0.8)]) # needs networkx sage: G.order(), G.size() # needs networkx (30, 52) sage: G.show() # long time # needs networkx sage.plot
- static RandomToleranceGraph(n, seed=None)#
Return a random tolerance graph.
The random tolerance graph is built from a random tolerance representation by using the function \(ToleranceGraph\). This representation is a list \(((l_0,r_0,t_0), (l_1,r_1,t_1), ..., (l_k,r_k,t_k))\) where \(k = n-1\) and \(I_i = (l_i,r_i)\) denotes a random interval and \(t_i\) a random positive value. The width of the representation is limited to \(n^2 * 2^n\).
Note
The vertices are named 0, 1, …, n-1. The tolerance representation used to create the graph is saved with the graph and can be recovered using
get_vertex()orget_vertices().INPUT:
n– number of vertices of the random graphseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
Every tolerance graph is perfect. Hence, the chromatic number is equal to the clique number
sage: g = graphs.RandomToleranceGraph(8) sage: g.clique_number() == g.chromatic_number() True
- static RandomTree(n, seed=None)#
Returns a random tree on \(n\) nodes numbered \(0\) through \(n-1\).
By Cayley’s theorem, there are \(n^{n-2}\) trees with vertex set \(\{0,1,...,n-1\}\). This constructor chooses one of these uniformly at random.
ALGORITHM:
The algorithm works by generating an \((n-2)\)-long random sequence of numbers chosen independently and uniformly from \(\{0,1,\ldots,n-1\}\) and then applies an inverse Prufer transformation.
INPUT:
n– number of vertices in the treeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
sage: G = graphs.RandomTree(10) sage: G.is_tree() True sage: G.show() # long time # needs sage.plot
- static RandomTreePowerlaw(n, gamma=3, tries=1000, seed=None)#
Return a tree with a power law degree distribution, or
Falseon failure.From the NetworkX documentation: a trial power law degree sequence is chosen and then elements are swapped with new elements from a power law distribution until the sequence makes a tree (size = order - 1).
INPUT:
n– number of verticesgamma– exponent of power law distributiontries– number of attempts to adjust sequence to make a treeseed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
EXAMPLES:
We check that the generated graph is a tree:
sage: G = graphs.RandomTreePowerlaw(10, 3) # needs networkx sage: G.is_tree() # needs networkx True sage: G.order(), G.size() # needs networkx (10, 9)
sage: G = graphs.RandomTreePowerlaw(15, 2) # needs networkx sage: if G: # random output # long time, needs networkx sage.plot ....: G.show()
- static RandomTriangulation(n, set_position=False, k=3, seed=None)#
Return a random inner triangulation of an outer face of degree
kwithnvertices in total.An inner triangulation is a plane graph all of whose faces (except the outer/unbounded face) are triangles (3-cycles).
INPUT:
n– the number of vertices of the graphk– the size of the outer faceset_position– boolean (defaultFalse); if set toTrue, this will compute coordinates for a planar drawing of the graph.seed– arandom.Randomseed or a Pythonintfor the random number generator (default:None)
OUTPUT:
A random graph chosen uniformly among the inner triangulations of a rooted \(k\)-gon with \(n\) vertices (including the \(k\) vertices from the outer face). This is a planar graph and comes with a combinatorial embedding. The vertices of the root edge are labelled
-1and-2and the outer face is the face returned byGraph.faces()in which-1and-2are consecutive vertices in this order.Because some triangulations have nontrivial automorphism groups, this may not be equal to the uniform distribution among inner triangulations of unrooted \(k\)-gons.
ALGORITHM:
The algorithm is taken from [PS2006], Section 5.
Starting from a planar \(k\)-gonal forest (represented by its contour as a sequence of vertices), one performs local closures, until no one is possible. A local closure amounts to replace in the cyclic contour word a sequence
in1, in2, in3, lf, in3byin1, in3.At every step of the algorithm, newly created edges are recorded in a graph, which will be returned at the end. The combinatorial embedding is also computed and recorded in the output graph.
See also
EXAMPLES:
sage: G = graphs.RandomTriangulation(6, True); G Graph on 6 vertices sage: G.is_planar() True sage: G.girth() 3 sage: G.plot(vertex_size=0, vertex_labels=False) # needs sage.plot Graphics object consisting of 13 graphics primitives sage: H = graphs.RandomTriangulation(7, k=5) sage: sorted(len(f) for f in H.faces()) [3, 3, 3, 3, 3, 3, 3, 5]
- static RandomUnitDiskGraph(n, radius=0.1, side=1, seed=None)#
Return a random unit disk graph of order \(n\).
A unit disk graph is the intersection graph of a family of unit disks in the Euclidean plane. That is a graph with one vertex per disk of the family and an edge between two vertices whenever they lie within a unit distance of each other. See the Wikipedia article Unit_disk_graph for more details.
INPUT:
n– number of nodesradius– float (default:0.1); two vertices at distance less thanradiusare connected by an edgeside– float (default:1); indicate the side of the area in which the points are drawnseed– seed of the random number generator
EXAMPLES:
When using twice the same seed, the vertices get the same positions:
sage: # needs scipy sage: from sage.misc.randstate import current_randstate sage: seed = current_randstate().seed() sage: G = graphs.RandomUnitDiskGraph(20, radius=.5, side=1, seed=seed) sage: H = graphs.RandomUnitDiskGraph(20, radius=.2, side=1, seed=seed) sage: H.is_subgraph(G, induced=False) True sage: H.size() <= G.size() True sage: Gpos = G.get_pos() sage: Hpos = H.get_pos() sage: all(Gpos[u] == Hpos[u] for u in G) True
When the radius is more than \(\sqrt{2 \text{side}}\), the graph is a clique:
sage: G = graphs.RandomUnitDiskGraph(10, radius=2, side=1) # needs scipy sage: G.is_clique() # needs scipy True
- static RingedTree(k, vertex_labels=True)#
Return the ringed tree on k-levels.
A ringed tree of level \(k\) is a binary tree with \(k\) levels (counting the root as a level), in which all vertices at the same level are connected by a ring.
More precisely, in each layer of the binary tree (i.e. a layer is the set of vertices \([2^i...2^{i+1}-1]\)) two vertices \(u,v\) are adjacent if \(u=v+1\) or if \(u=2^i\) and \(v=`2^{i+1}-1\).
Ringed trees are defined in [CFHM2013].
INPUT:
k– the number of levels of the ringed tree.vertex_labels(boolean) – whether to label vertices as binary words (default) or as integers.
EXAMPLES:
sage: # needs networkx sage: G = graphs.RingedTree(5) sage: P = G.plot(vertex_labels=False, vertex_size=10) # needs sage.plot sage: P.show() # long time # needs sage.plot sage: G.vertices(sort=True) ['', '0', '00', '000', '0000', '0001', '001', '0010', '0011', '01', '010', '0100', '0101', '011', '0110', '0111', '1', '10', '100', '1000', '1001', '101', '1010', '1011', '11', '110', '1100', '1101', '111', '1110', '1111']
- static RobertsonGraph()#
Return the Robertson graph.
See the Wikipedia article Robertson_graph.
EXAMPLES:
sage: # needs networkx sage: g = graphs.RobertsonGraph() sage: g.order() 19 sage: g.size() 38 sage: g.diameter() 3 sage: g.girth() 5 sage: g.charpoly().factor() (x - 4) * (x - 1)^2 * (x^2 + x - 5) * (x^2 + x - 1) * (x^2 - 3)^2 * (x^2 + x - 4)^2 * (x^2 + x - 3)^2 sage: g.chromatic_number() 3 sage: g.is_hamiltonian() # needs sage.numerical.mip True sage: g.is_vertex_transitive() False
- static RookGraph(dim_list, radius=None, relabel=False)#
Return the \(d\)-dimensional Rook’s Graph with prescribed dimensions.
The 2-dimensional Rook’s Graph of parameters \(n\) and \(m\) is a graph with \(nm\) vertices in which each vertex represents a square in an \(n \times m\) chessboard, and each edge corresponds to a legal move by a rook.
The \(d\)-dimensional Rook Graph with \(d >= 2\) has for vertex set the cells of a \(d\)-dimensional grid with prescribed dimensions, and each edge corresponds to a legal move by a rook in any of the dimensions.
The Rook’s Graph for an \(n\times m\) chessboard may also be defined as the Cartesian product of two complete graphs \(K_n \square K_m\).
INPUT:
dim_list– iterable (list, set, dict); provides the dimensions \(n_1, n_2, \ldots, n_d\), with \(n_i \geq 1\), of the chessboardradius– integer (default:None); by setting the radius to a positive integer, one may decrease the power of the rook to at mostradiussteps. When the radius is 1, the resulting graph is a \(d\)-dimensional grid.relabel– boolean (default:False); indicates whether the vertices must be relabeled as integers
EXAMPLES:
The \((n,m)\)-Rook’s Graph is isomorphic to the Cartesian product of two complete graphs:
sage: G = graphs.RookGraph( [3, 4] ) sage: H = ( graphs.CompleteGraph(3) ).cartesian_product( graphs.CompleteGraph(4) ) sage: G.is_isomorphic( H ) True
When the radius is 1, the Rook’s Graph is a grid:
sage: G = graphs.RookGraph( [3, 3, 4], radius=1 ) sage: H = graphs.GridGraph( [3, 4, 3] ) sage: G.is_isomorphic( H ) True
- static RoseWindowGraph(n, a, r)#
Return a rose window graph with \(2n\) nodes.
The rose window graphs is a family of tetravalant graphs introduced in [Wilson2008]. The parameters \(n\), \(a\) and \(r\) are integers such that \(n > 2\), \(1 \leq a, r < n\), and \(r \neq n / 2\).
INPUT:
n– the number of nodes is \(2 * n\)a– integer such that \(1 \leq a < n\) determining a-spoke edgesr– integer such that \(1 \leq r < n\) and \(r \neq n / 2\) determining how inner vertices are connected
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the rose window graphs are displayed as an inner and outer cycle pair, with the first n nodes drawn on the outer circle. The first (0) node is drawn at the top of the outer-circle, moving counterclockwise after that. The inner circle is drawn with the (n)th node at the top, then counterclockwise as well. Vertices in the outer circle are connected in the circular manner, vertices in the inner circle are connected when their label have difference \(r\) (mod n). Vertices on the outer rim are connected with the vertices on the inner rim when they are at the same position and when they are \(a\) apart.
EXAMPLES:
The vertices of a rose window graph have all degree 4:
sage: G = graphs.RoseWindowGraph(5, 1, 2) sage: all(G.degree(u) == 4 for u in G) True
The smallest rose window graph as parameters \((3, 2, 1)\):
sage: G = graphs.RoseWindowGraph(3, 2, 1) sage: all(G.degree(u) == 4 for u in G) True
- static SchlaefliGraph()#
Return the Schläfli graph.
The Schläfli graph is the only strongly regular graphs of parameters \((27,16,10,8)\) (see [GR2001]).
For more information, see the Wikipedia article Schläfli_graph.
See also
Graph.is_strongly_regular()– tests whether a graph is strongly regular and/or returns its parameters.Todo
Find a beautiful layout for this beautiful graph.
EXAMPLES:
Checking that the method actually returns the Schläfli graph:
sage: S = graphs.SchlaefliGraph() sage: S.is_strongly_regular(parameters=True) (27, 16, 10, 8)
The graph is vertex-transitive:
sage: S.is_vertex_transitive() # needs sage.groups True
The neighborhood of each vertex is isomorphic to the complement of the Clebsch graph:
sage: neighborhood = S.subgraph(vertices=S.neighbors(0)) sage: graphs.ClebschGraph().complement().is_isomorphic(neighborhood) True
- static ShrikhandeGraph()#
Return the Shrikhande graph.
For more information, see the MathWorld article on the Shrikhande graph or the Wikipedia article Shrikhande_graph.
See also
Graph.is_strongly_regular()– tests whether a graph is strongly regular and/or returns its parameters.EXAMPLES:
The Shrikhande graph was defined by S. S. Shrikhande in 1959. It has \(16\) vertices and \(48\) edges, and is strongly regular of degree \(6\) with parameters \((2,2)\):
sage: G = graphs.ShrikhandeGraph(); G Shrikhande graph: Graph on 16 vertices sage: G.order() 16 sage: G.size() 48 sage: G.is_regular(6) True sage: set([ len([x for x in G.neighbors(i) if x in G.neighbors(j)]) ....: for i in range(G.order()) ....: for j in range(i) ]) {2}
It is non-planar, and both Hamiltonian and Eulerian:
sage: G.is_planar() False sage: G.is_hamiltonian() # needs sage.numerical.mip True sage: G.is_eulerian() True
It has radius \(2\), diameter \(2\), and girth \(3\):
sage: G.radius() 2 sage: G.diameter() 2 sage: G.girth() 3
Its chromatic number is \(4\) and its automorphism group is of order \(192\):
sage: G.chromatic_number() 4 sage: G.automorphism_group().cardinality() # needs sage.groups 192
It is an integral graph since it has only integral eigenvalues:
sage: G.characteristic_polynomial().factor() # needs sage.libs.pari sage.modules (x - 6) * (x - 2)^6 * (x + 2)^9
It is a toroidal graph, and its embedding on a torus is dual to an embedding of the Dyck graph (
DyckGraph).
- static SierpinskiGasketGraph(n)#
Return the Sierpinski Gasket graph of generation \(n\).
All vertices but 3 have valence 4.
INPUT:
\(n\) – an integer
OUTPUT:
a graph \(S_n\) with \(3 (3^{n-1}+1)/2\) vertices and \(3^n\) edges, closely related to the famous Sierpinski triangle fractal.
All these graphs have a triangular shape, and three special vertices at top, bottom left and bottom right. These are the only vertices of valence 2, all the other ones having valence 4.
The graph \(S_1\) (generation \(1\)) is a triangle.
The graph \(S_{n+1}\) is obtained from the disjoint union of three copies A,B,C of \(S_n\) by identifying pairs of vertices: the top vertex of A with the bottom left vertex of B, the bottom right vertex of B with the top vertex of C, and the bottom left vertex of C with the bottom right vertex of A.
See also
HanoiTowerGraph(). There is another family of graphs called Sierpinski graphs, where all vertices but 3 have valence 3. They are available usinggraphs.HanoiTowerGraph(3, n).
EXAMPLES:
sage: # needs sage.modules sage: s4 = graphs.SierpinskiGasketGraph(4); s4 Graph on 42 vertices sage: s4.size() 81 sage: s4.degree_histogram() [0, 0, 3, 0, 39] sage: s4.is_hamiltonian() True
REFERENCES:
- static SimsGewirtzGraph()#
Return the Sims-Gewirtz Graph.
This graph is obtained from the Higman Sims graph by considering the graph induced by the vertices at distance two from the vertices of an (any) edge. It is the only strongly regular graph with parameters \(v = 56\), \(k = 10\), \(\lambda = 0\), \(\mu = 2\)
For more information on the Sylvester graph, see https://www.win.tue.nl/~aeb/graphs/Sims-Gewirtz.html or its Wikipedia article Gewirtz_graph.
See also
EXAMPLES:
sage: g = graphs.SimsGewirtzGraph(); g Sims-Gewirtz Graph: Graph on 56 vertices sage: g.order() 56 sage: g.size() 280 sage: g.is_strongly_regular(parameters = True) (56, 10, 0, 2)
- static SousselierGraph()#
Return the Sousselier Graph.
The Sousselier graph is a hypohamiltonian graph on 16 vertices and 27 edges. For more information, see Wikipedia article Sousselier_graph or the corresponding French Wikipedia page.
EXAMPLES:
sage: g = graphs.SousselierGraph() sage: g.order() 16 sage: g.size() 27 sage: g.radius() 2 sage: g.diameter() 3 sage: g.automorphism_group().cardinality() # needs sage.groups 2 sage: g.is_hamiltonian() # needs sage.numerical.mip False sage: g.delete_vertex(g.random_vertex()) sage: g.is_hamiltonian() # needs sage.numerical.mip True
- static SquaredSkewHadamardMatrixGraph(n)#
Pseudo-\(OA(2n,4n-1)\)-graph from a skew Hadamard matrix of order \(4n\)
A strongly regular graph with parameters of the orthogonal array graph
OrthogonalArrayBlockGraph(), also known as pseudo Latin squares graph \(L_{2n}(4n-1)\), constructed from a skew Hadamard matrix of order \(4n\), due to Goethals and Seidel, see [BL1984].See also
EXAMPLES:
sage: # needs sage.combinat sage.modules sage: G = graphs.SquaredSkewHadamardMatrixGraph(4) sage: G.is_strongly_regular(parameters=True) (225, 112, 55, 56) sage: G = graphs.SquaredSkewHadamardMatrixGraph(5) sage: G.is_strongly_regular(parameters=True) # long time (361, 180, 89, 90) sage: G = graphs.SquaredSkewHadamardMatrixGraph(9) sage: G.is_strongly_regular(parameters=True) # not tested (1225, 612, 305, 306)
- static StarGraph(n)#
Return a star graph with \(n + 1\) nodes.
A Star graph is a basic structure where one node is connected to all other nodes.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each star graph will be displayed with the first (0) node in the center, the second node (1) at the top, with the rest following in a counterclockwise manner. (0) is the node connected to all other nodes.
The star graph is a good opportunity to compare efficiency of filling a position dictionary vs. using the spring-layout algorithm for plotting. As far as display, the spring-layout should push all other nodes away from the (0) node, and thus look very similar to this constructor’s positioning.
EXAMPLES:
sage: import networkx # needs networkx
Compare the plots:
sage: # needs networkx sage.plot sage: n = networkx.star_graph(23) sage: spring23 = Graph(n) sage: posdict23 = graphs.StarGraph(23) sage: spring23.show() # long time sage: posdict23.show() # long time
View many star graphs as a Sage Graphics Array
With this constructor (i.e., the position dictionary filled)
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.StarGraph(i+3) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
Compared to plotting with the spring-layout algorithm
sage: # needs networkx sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: spr = networkx.star_graph(i+3) ....: k = Graph(spr) ....: g.append(k) sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static SuzukiGraph()#
Return the Suzuki Graph.
The Suzuki graph has 1782 vertices, and is strongly regular with parameters \((1782,416,100,96)\). Known as S.15 in [Hub1975].
Note
It takes approximately 50 seconds to build this graph. Do not be too impatient.
EXAMPLES:
sage: g = graphs.SuzukiGraph(); g # optional internet # not tested Suzuki graph: Graph on 1782 vertices sage: g.is_strongly_regular(parameters=True) # optional internet # not tested (1782, 416, 100, 96)
- static SwitchedSquaredSkewHadamardMatrixGraph(n)#
A strongly regular graph in Seidel switching class of \(SquaredSkewHadamardMatrixGraph\)
A strongly regular graph in the
Seidel switchingclass of the disjoint union of a 1-vertex graph and the one produced byPseudo-L_{2n}(4n-1)In this case, the other possible parameter set of a strongly regular graph in the Seidel switching class of the latter graph (see [BH2012]) coincides with the set of parameters of the complement of the graph returned by this function.
See also
EXAMPLES:
sage: # needs sage.combinat sage.modules sage: g = graphs.SwitchedSquaredSkewHadamardMatrixGraph(4) sage: g.is_strongly_regular(parameters=True) (226, 105, 48, 49) sage: from sage.combinat.designs.twographs import twograph_descendant sage: twograph_descendant(g, 0).is_strongly_regular(parameters=True) (225, 112, 55, 56) sage: gc = g.complement() sage: twograph_descendant(gc, 0).is_strongly_regular(parameters=True) (225, 112, 55, 56)
- static SylvesterGraph()#
Return the Sylvester Graph.
This graph is obtained from the Hoffman Singleton graph by considering the graph induced by the vertices at distance two from the vertices of an (any) edge.
For more information on the Sylvester graph, see https://www.win.tue.nl/~aeb/graphs/Sylvester.html.
See also
EXAMPLES:
sage: g = graphs.SylvesterGraph(); g Sylvester Graph: Graph on 36 vertices sage: g.order() 36 sage: g.size() 90 sage: g.is_regular(k=5) True
- static SymplecticDualPolarGraph(m, q)#
Return the Symplectic Dual Polar Graph \(DSp(m,q)\).
For more information on Symplectic Dual Polar graphs, see [BCN1989] and Sect. 2.3.1 of [Coh1981].
INPUT:
m,q– integers; \(q\) must be a prime power, and \(m\) must be even
EXAMPLES:
sage: G = graphs.SymplecticDualPolarGraph(6,3); G # not tested (long time) Symplectic Dual Polar Graph DSp(6, 3): Graph on 1120 vertices sage: G.is_distance_regular(parameters=True) # not tested (long time) ([39, 36, 27, None], [None, 1, 4, 13])
- static SymplecticPolarGraph(d, q, algorithm=None)#
Return the Symplectic Polar Graph \(Sp(d,q)\).
The Symplectic Polar Graph \(Sp(d,q)\) is built from a projective space of dimension \(d-1\) over a field \(F_q\), and a symplectic form \(f\). Two vertices \(u,v\) are made adjacent if \(f(u,v)=0\).
See the page on symplectic graphs on Andries Brouwer’s website.
INPUT:
d,q– integers; note that only even values of \(d\) are accepted by the function.algorithm– string (default:None); if set to'gap', then the computation is carried via GAP library interface, computing totally singular subspaces, which is faster for \(q>3\). Otherwise it is done directly.
EXAMPLES:
Computation of the spectrum of \(Sp(6,2)\):
sage: g = graphs.SymplecticPolarGraph(6, 2) sage: g.is_strongly_regular(parameters=True) (63, 30, 13, 15) sage: set(g.spectrum()) == {-5, 3, 30} # needs sage.rings.number_field True
The parameters of \(Sp(4,q)\) are the same as of \(O(5,q)\), but they are not isomorphic if \(q\) is odd:
sage: G = graphs.SymplecticPolarGraph(4, 3) sage: G.is_strongly_regular(parameters=True) (40, 12, 2, 4) sage: # needs sage.libs.gap sage: O = graphs.OrthogonalPolarGraph(5, 3) sage: O.is_strongly_regular(parameters=True) (40, 12, 2, 4) sage: O.is_isomorphic(G) False sage: S = graphs.SymplecticPolarGraph(6, 4, algorithm="gap") # not tested (long time) sage: S.is_strongly_regular(parameters=True) # not tested (long time) (1365, 340, 83, 85)
- static SzekeresSnarkGraph()#
Return the Szekeres Snark Graph.
The Szekeres graph is a snark with 50 vertices and 75 edges. For more information on this graph, see the Wikipedia article Szekeres_snark.
EXAMPLES:
sage: g = graphs.SzekeresSnarkGraph() sage: g.order() 50 sage: g.size() 75 sage: g.chromatic_number() 3
- static T2starGeneralizedQuadrangleGraph(q, dual=False, hyperoval=None, field=None, check_hyperoval=True)#
Return the collinearity graph of the generalized quadrangle \(T_2^*(q)\), or of its dual
Let \(q=2^k\) and \(\Theta=PG(3,q)\). \(T_2^*(q)\) is a generalized quadrangle (Wikipedia article Generalized_quadrangle) of order \((q-1,q+1)\), see 3.1.3 in [PT2009]. Fix a plane \(\Pi \subset \Theta\) and a hyperoval \(O \subset \Pi\). The points of \(T_2^*(q):=T_2^*(O)\) are the points of \(\Theta\) outside \(\Pi\), and the lines are the lines of \(\Theta\) outside \(\Pi\) that meet \(\Pi\) in a point of \(O\).
INPUT:
q– a power of twodual– boolean (default:False); whether to return the graph of \(T_2^*(O)\) or of the dual \(T_2^*(O)\) (whenTrue)hyperoval– a hyperoval (i.e. a complete 2-arc; a set of points in the plane meeting every line in 0 or 2 points) in the plane of points with 0th coordinate 0 in \(PG(3,q)\) over the fieldfield. Each point ofhyperovalmust be a length 4 vector overfieldwith 1st non-0 coordinate equal to 1. By default,hyperovalandfieldare not specified, and constructed on the fly. In particular,hyperovalwe build is the classical one, i.e. a conic with the point of intersection of its tangent lines.field– an instance of a finite field of order \(q\), must be provided ifhyperovalis providedcheck_hyperoval– boolean (default:True); whether to checkhyperovalfor correctness or not
EXAMPLES:
using the built-in construction:
sage: # needs sage.combinat sage.rings.finite_rings sage: g = graphs.T2starGeneralizedQuadrangleGraph(4); g T2*(O,4); GQ(3, 5): Graph on 64 vertices sage: g.is_strongly_regular(parameters=True) (64, 18, 2, 6) sage: g = graphs.T2starGeneralizedQuadrangleGraph(4, dual=True); g T2*(O,4)*; GQ(5, 3): Graph on 96 vertices sage: g.is_strongly_regular(parameters=True) (96, 20, 4, 4)
supplying your own hyperoval:
sage: # needs sage.combinat sage.rings.finite_rings sage: F = GF(4,'b') sage: O = [vector(F,(0,0,0,1)),vector(F,(0,0,1,0))] + [vector(F, (0,1,x^2,x)) ....: for x in F] sage: g = graphs.T2starGeneralizedQuadrangleGraph(4, hyperoval=O, field=F); g T2*(O,4); GQ(3, 5): Graph on 64 vertices sage: g.is_strongly_regular(parameters=True) (64, 18, 2, 6)
- static TabacjnGraph(n, a, b, r)#
Return a Tabačjn graph with \(2n\) nodes.
The Tabačjn graphs is a family of pentavalent bicirculants graphs proposed in [AHKOS2014] as a generalization of generalized Petersen graphs. The parameters \(n\), \(a\), \(b\), \(r\) are integers such that \(n \geq 3\), \(1 \leq a, b, r \leq n - 1\), with \(a \neq b\) and \(r \neq n / 2\).
INPUT:
n– the number of nodes is \(2 * n\)a– integer such that \(0 < a < n\) and \(a \neq b\), that determines a-spoke edgesb– integer such that \(0 < b < n\) and \(b \neq a\), that determines b-spoke edgesr– integer such that \(0 < r < n\) and \(r \neq n/2\) determining how inner vertices are connected
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the rose window graphs are displayed as an inner and outer cycle pair, with the first n nodes drawn on the outer circle. The first (0) node is drawn at the top of the outer-circle, moving counterclockwise after that. The inner circle is drawn with the (n)th node at the top, then counterclockwise as well. Vertices in the outer circle are connected in the circular manner, vertices in the inner circle are connected when their label have difference \(r\) (mod n). Vertices on the outer rim are connected with the vertices on the inner rim when they are at the same position and when they are \(a\) and \(b\) apart.
EXAMPLES:
sage: G = graphs.TabacjnGraph(3, 1, 2, 1) sage: G.degree() [5, 5, 5, 5, 5, 5] sage: G.is_isomorphic(graphs.CompleteGraph(6)) True sage: G = graphs.TabacjnGraph(6, 1, 5, 2) sage: I = graphs.IcosahedralGraph() sage: G.is_isomorphic(I) True
- static TadpoleGraph(n1, n2)#
Return a tadpole graph with n1+n2 nodes.
A tadpole graph is a path graph (order n2) connected to a cycle graph (order n1).
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, the cycle graph will be drawn in the lower-left corner with the (n1)th node at a 45 degree angle above the right horizontal center of the cycle graph, leading directly into the path graph.
EXAMPLES:
Construct and show a tadpole graph Cycle = 13, Stick = 4:
sage: g = graphs.TadpoleGraph(13, 4); g Tadpole graph: Graph on 17 vertices sage: g.show() # long time # needs sage.plot
- static TaylorTwographDescendantSRG(q, clique_partition=False)#
Return the descendant graph of the Taylor’s two-graph for \(U_3(q)\), \(q\) odd.
This is a strongly regular graph with parameters \((v,k,\lambda,\mu)=(q^3, (q^2+1)(q-1)/2, (q-1)^3/4-1, (q^2+1)(q-1)/4)\) obtained as a two-graph descendant of the
Taylor's two-graph\(T\). This graph admits a partition into cliques of size \(q\), which are useful inTaylorTwographSRG(), a strongly regular graph on \(q^3+1\) vertices in the Seidel switching class of \(T\), for which we need \((q^2+1)/2\) cliques. The cliques are the \(q^2\) lines on \(v_0\) of the projective plane containing the unital for \(U_3(q)\), and intersecting the unital (i.e. the vertices of the graph and the point we remove) in \(q+1\) points. This is all taken from §7E of [BL1984].INPUT:
q– a power of an odd prime numberclique_partition– boolean (default:False); when set toTrue, return \(q^2-1\) cliques of size \(q\) with empty pairwise intersection. (Removing all of them leaves a clique, too), and the point removed from the unital.
EXAMPLES:
sage: # needs sage.rings.finite_rings sage: g = graphs.TaylorTwographDescendantSRG(3); g Taylor two-graph descendant SRG: Graph on 27 vertices sage: g.is_strongly_regular(parameters=True) (27, 10, 1, 5) sage: from sage.combinat.designs.twographs import taylor_twograph sage: T = taylor_twograph(3) # long time sage: g.is_isomorphic(T.descendant(T.ground_set()[1])) # long time True sage: g = graphs.TaylorTwographDescendantSRG(5) # not tested (long time) sage: g.is_strongly_regular(parameters=True) # not tested (long time) (125, 52, 15, 26)
- static TaylorTwographSRG(q)#
Return a strongly regular graph from the Taylor’s two-graph for \(U_3(q)\), \(q\) odd
This is a strongly regular graph with parameters \((v,k,\lambda,\mu)=(q^3+1, q(q^2+1)/2, (q^2+3)(q-1)/4, (q^2+1)(q+1)/4)\) in the Seidel switching class of
Taylor two-graph. Details are in §7E of [BL1984].INPUT:
q– a power of an odd prime number
See also
EXAMPLES:
sage: t = graphs.TaylorTwographSRG(3); t # needs sage.rings.finite_rings Taylor two-graph SRG: Graph on 28 vertices sage: t.is_strongly_regular(parameters=True) # needs sage.rings.finite_rings (28, 15, 6, 10)
- static TetrahedralGraph()#
Return a tetrahedral graph (with 4 nodes).
A tetrahedron is a 4-sided triangular pyramid. The tetrahedral graph corresponds to the connectivity of the vertices of the tetrahedron. This graph is equivalent to a wheel graph with 4 nodes and also a complete graph on four nodes. (See examples below).
PLOTTING: The Tetrahedral graph should be viewed in 3 dimensions. We choose to use a planar embedding of the graph. We hope to add rotatable, 3-dimensional viewing in the future. In such a case, a argument will be added to select the desired layout.
EXAMPLES:
Construct and show a Tetrahedral graph:
sage: g = graphs.TetrahedralGraph() sage: g.show() # long time # needs sage.plot
The following example requires networkx:
sage: import networkx as NX # needs networkx
Compare this Tetrahedral, Wheel(4), Complete(4), and the Tetrahedral plotted with the spring-layout algorithm below in a Sage graphics array:
sage: # needs networkx sage.plot sage: tetra_pos = graphs.TetrahedralGraph() sage: tetra_spring = Graph(NX.tetrahedral_graph()) sage: wheel = graphs.WheelGraph(4) sage: complete = graphs.CompleteGraph(4) sage: g = [tetra_pos, tetra_spring, wheel, complete] sage: j = [] sage: for i in range(2): ....: n = [] ....: for m in range(2): ....: n.append(g[i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) sage: G = graphics_array(j) sage: G.show() # long time
- static ThomsenGraph()#
Return the Thomsen Graph.
The Thomsen Graph is actually a complete bipartite graph with \((n1, n2) = (3, 3)\). It is also called the Utility graph.
PLOTTING: See CompleteBipartiteGraph.
EXAMPLES:
sage: T = graphs.ThomsenGraph() sage: T Thomsen graph: Graph on 6 vertices sage: T.graph6_string() 'EFz_' sage: (graphs.ThomsenGraph()).show() # long time # needs sage.plot
- static TietzeGraph()#
Return the Tietze Graph.
For more information on the Tietze Graph, see the Wikipedia article Tietze%27s_graph.
EXAMPLES:
sage: g = graphs.TietzeGraph() sage: g.order() 12 sage: g.size() 18 sage: g.diameter() 3 sage: g.girth() 3 sage: g.automorphism_group().cardinality() # needs sage.groups 12 sage: g.automorphism_group().is_isomorphic(groups.permutation.Dihedral(6)) # needs sage.groups True
- static ToleranceGraph(tolrep)#
Return the graph generated by the tolerance representation
tolrep.The tolerance representation
tolrepis described by the list \(((l_0,r_0,t_0), (l_1,r_1,t_1), \ldots, (l_k,r_k,t_k))\) where \(I_i = (l_i,r_i)\) denotes a closed interval on the real line with \(l_i < r_i\) and \(t_i\) a strictly positive value, called tolerance. This representation generates the tolerance graph with the vertex set \(\{0,1, \ldots, k\}\) and the edge set \(\{(i,j): |I_i \cap I_j| \ge \min\{t_i, t_j\}\}\) where \(|I_i \cap I_j|\) denotes the length of the intersection of \(I_i\) and \(I_j\).INPUT:
tolrep– list of triples \((l_i,r_i,t_i)\) where \((l_i,r_i)\) denotes a closed interval on the real line and \(t_i\) a positive value.
Note
The vertices are named \(0, 1, \ldots, k\). The tolerance representation used to create the graph is saved with the graph and can be recovered using
get_vertex()orget_vertices().EXAMPLES:
The following code creates a tolerance representation
tolrep, generates its tolerance graphg, and applies some checks:sage: tolrep = [(1,4,3),(1,2,1),(2,3,1),(0,3,3)] sage: g = graphs.ToleranceGraph(tolrep) sage: g.get_vertex(3) (0, 3, 3) sage: neigh = g.neighbors(3) sage: for v in neigh: print(g.get_vertex(v)) (1, 2, 1) (2, 3, 1) sage: g.is_interval() False sage: g.is_weakly_chordal() True
The intervals in the list need not be distinct
sage: tolrep2 = [(0,4,5),(1,2,1),(2,3,1),(0,4,5)] sage: g2 = graphs.ToleranceGraph(tolrep2) sage: g2.get_vertices() {0: (0, 4, 5), 1: (1, 2, 1), 2: (2, 3, 1), 3: (0, 4, 5)} sage: g2.is_isomorphic(g) True
Real values are also allowed
sage: tolrep = [(0.1,3.3,4.4),(1.1,2.5,1.1),(1.4,4.4,3.3)] sage: g = graphs.ToleranceGraph(tolrep) sage: g.is_isomorphic(graphs.PathGraph(3)) True
- static Toroidal6RegularGrid2dGraph(p, q)#
Return a toroidal 6-regular grid.
The toroidal 6-regular grid is a 6-regular graph on \(p \times q\) vertices and its elements have coordinates \((i, j)\) for \(i \in \{0...p-1\}\) and \(j \in \{0...q-1\}\).
Its edges are those of the
ToroidalGrid2dGraph(), to which are added the edges between \((i, j)\) and \(((i + 1) \% p, (j + 1) \% q)\).INPUT:
p, q– integers (see above)
EXAMPLES:
The toroidal 6-regular grid on \(25\) elements:
sage: g = graphs.Toroidal6RegularGrid2dGraph(5,5) sage: g.is_regular(k=6) True sage: g.is_vertex_transitive() # needs sage.groups True sage: g.line_graph().is_vertex_transitive() # needs sage.groups True sage: g.automorphism_group().cardinality() # needs sage.groups 300 sage: g.is_hamiltonian() # needs sage.numerical.mip True
- static ToroidalGrid2dGraph(p, q)#
Return a toroidal 2-dimensional grid graph with \(p \times q\) nodes (\(p\) rows and \(q\) columns).
The toroidal 2-dimensional grid with parameters \(p,q\) is the 2-dimensional grid graph with identical parameters to which are added the edges \(((i, 0), (i, q - 1))\) and \(((0, i), (p - 1, i))\).
EXAMPLES:
The toroidal 2-dimensional grid is a regular graph, while the usual 2-dimensional grid is not
sage: tgrid = graphs.ToroidalGrid2dGraph(8,9) sage: print(tgrid) Toroidal 2D Grid Graph with parameters 8,9 sage: grid = graphs.Grid2dGraph(8,9) sage: grid.is_regular() False sage: tgrid.is_regular() True
- static TruncatedIcosidodecahedralGraph()#
Return the truncated icosidodecahedron.
The truncated icosidodecahedron is an Archimedean solid with 30 square faces, 20 regular hexagonal faces, 12 regular decagonal faces, 120 vertices and 180 edges. For more information, see the Wikipedia article Truncated_icosidodecahedron.
EXAMPLES:
Unfortunately, this graph can not be constructed currently, due to numerical issues:
sage: g = graphs.TruncatedIcosidodecahedralGraph(); g # needs sage.geometry.polyhedron sage.groups sage.rings.number_field Traceback (most recent call last): ... ValueError: *Error: Numerical inconsistency is found. Use the GMP exact arithmetic. sage: g.order(), g.size() # not tested # needs sage.geometry.polyhedron sage.groups sage.rings.number_field (120, 180)
- static TruncatedTetrahedralGraph()#
Return the truncated tetrahedron.
The truncated tetrahedron is an Archimedean solid with 12 vertices and 18 edges. For more information, see the Wikipedia article Truncated_tetrahedron.
EXAMPLES:
sage: g = graphs.TruncatedTetrahedralGraph(); g Truncated Tetrahedron: Graph on 12 vertices sage: g.order(), g.size() (12, 18) sage: g.is_isomorphic(polytopes.simplex(3).truncation().graph()) # needs sage.geometry.polyhedron True
- static TruncatedWittGraph()#
Return the truncated Witt graph.
This builds the large Witt graph, then removes all vertices whose codeword start with a 1.
The graph is distance-regular with intersection array \([15,14,12;1,1,9]\).
EXAMPLES:
sage: # long time, needs sage.libs.pari sage.modules sage: G = graphs.TruncatedWittGraph() sage: G.is_distance_regular(True) ([15, 14, 12, None], [None, 1, 1, 9])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 367.
- static TuranGraph(n, r)#
Returns the Turan graph with parameters \(n, r\).
Turan graphs are complete multipartite graphs with \(n\) vertices and \(r\) subsets, denoted \(T(n,r)\), with the property that the sizes of the subsets are as close to equal as possible. The graph \(T(n,r)\) will have \(n \pmod r\) subsets of size \(\lfloor n/r \rfloor\) and \(r - (n \pmod r)\) subsets of size \(\lceil n/r \rceil\). See the Wikipedia article Turan_graph for more information.
INPUT:
n– integer; the number of vertices in the graphr– integer; the number of partitions of the graph
EXAMPLES:
The Turan graph is a complete multipartite graph:
sage: g = graphs.TuranGraph(13, 4) sage: k = graphs.CompleteMultipartiteGraph([3,3,3,4]) sage: g.is_isomorphic(k) True
The Turan graph \(T(n,r)\) has \(\frac{(r-1)(n^2-s^2)}{2r} + \frac{s(s-1)}{2}\) edges, where \(s = n \mod r\) (github issue #34249):
sage: n = 12 sage: r = 8 sage: g = graphs.TuranGraph(n, r) sage: def count(n, r): ....: s = n % r ....: return (r - 1) * (n**2 - s**2) / (2*r) + s*(s - 1)/2 sage: g.size() == count(n, r) True sage: n = randint(3, 100) sage: r = randint(2, n - 1) sage: g = graphs.TuranGraph(n, r) sage: g.size() == count(n, r) True
- static Tutte12Cage()#
Return the Tutte 12-Cage.
See the Wikipedia article Tutte_12-cage.
EXAMPLES:
sage: # needs networkx sage: g = graphs.Tutte12Cage() sage: g.order() 126 sage: g.size() 189 sage: g.girth() 12 sage: g.diameter() 6 sage: g.show() # needs sage.plot
- static TutteCoxeterGraph(embedding=2)#
Return the Tutte-Coxeter graph.
See the Wikipedia article Tutte-Coxeter_graph.
INPUT:
embedding– integer (default:2); two embeddings are available, and can be selected by settingembeddingto 1 or 2
EXAMPLES:
sage: # needs networkx sage: g = graphs.TutteCoxeterGraph() sage: g.order() 30 sage: g.size() 45 sage: g.girth() 8 sage: g.diameter() 4 sage: g.show() # needs sage.plot sage: graphs.TutteCoxeterGraph(embedding=1).show() # long time # needs sage.plot
- static TutteGraph()#
Return the Tutte Graph.
The Tutte graph is a 3-regular, 3-connected, and planar non-hamiltonian graph. For more information on the Tutte Graph, see the Wikipedia article Tutte_graph.
EXAMPLES:
sage: g = graphs.TutteGraph() sage: g.order() 46 sage: g.size() 69 sage: g.is_planar() True sage: g.vertex_connectivity() # long time # needs sage.numerical.mip 3 sage: g.girth() 4 sage: g.automorphism_group().cardinality() # needs sage.groups 3 sage: g.is_hamiltonian() # needs sage.numerical.mip False
- static U42Graph216()#
Return a (216,40,4,8)-strongly regular graph from [CRS2016].
Build the graph, interpreting the \(U_4(2)\)-action considered in [CRS2016] as the one on the hyperbolic lines of the corresponding unitary polar space, and then doing the unique merging of the orbitals leading to a graph with the parameters in question.
EXAMPLES:
sage: G=graphs.U42Graph216() # optional - gap_package_grape sage: G.is_strongly_regular(parameters=True) # optional - gap_package_grape (216, 40, 4, 8)
- static U42Graph540()#
Return a (540,187,58,68)-strongly regular graph from [CRS2016].
Build the graph, interpreting the \(U_4(2)\)-action considered in [CRS2016] as the action of \(U_4(2)=Sp_4(3)<U_4(3)\) on the nonsingular, w.r.t. to the Hermitean form stabilised by \(U_4(3)\), points of the 3-dimensional projective space over \(GF(9)\). There are several possible mergings of orbitals, some leading to non-isomorphic graphs with the same parameters. We found the merging here using [FK1991].
EXAMPLES:
sage: G = graphs.U42Graph540() # optional - gap_package_grape sage: G.is_strongly_regular(parameters=True) # optional - gap_package_grape (540, 187, 58, 68)
- static USAMap(continental=False)#
Return states of USA as a graph of common border.
The graph has an edge between those states that have common land border line or point. Hence for example Colorado and Arizona are marked as neighbors, but Michigan and Minnesota are not.
INPUT:
continental– boolean (default:False); whether to exclude Alaska and Hawaii
EXAMPLES:
How many states are neighbor’s neighbor for Pennsylvania:
sage: USA = graphs.USAMap() sage: distance = USA.shortest_path_lengths('Pennsylvania') sage: len([n2 for n2, d in distance.items() if d == 2]) 7
Diameter for continental USA:
sage: USAcont = graphs.USAMap(continental=True) sage: USAcont.diameter() 11
- static UnitaryDualPolarGraph(m, q)#
Return the Dual Unitary Polar Graph \(U(m,q)\).
For more information on Unitary Dual Polar graphs, see [BCN1989] and Sect. 2.3.1 of [Coh1981].
INPUT:
m,q– integers; \(q\) must be a prime power
EXAMPLES:
The point graph of a generalized quadrangle (see Wikipedia article Generalized_quadrangle, [PT2009]) of order (8,4):
sage: G = graphs.UnitaryDualPolarGraph(5,2); G # long time # needs sage.libs.gap Unitary Dual Polar Graph DU(5, 2); GQ(8, 4): Graph on 297 vertices sage: G.is_strongly_regular(parameters=True) # long time # needs sage.libs.gap (297, 40, 7, 5)
Another way to get the generalized quadrangle of order (2,4):
sage: G = graphs.UnitaryDualPolarGraph(4,2); G # needs sage.libs.gap Unitary Dual Polar Graph DU(4, 2); GQ(2, 4): Graph on 27 vertices sage: G.is_isomorphic(graphs.OrthogonalPolarGraph(6,2,'-')) # needs sage.libs.gap True
A bigger graph:
sage: G = graphs.UnitaryDualPolarGraph(6,2); G # not tested (long time) Unitary Dual Polar Graph DU(6, 2): Graph on 891 vertices sage: G.is_distance_regular(parameters=True) # not tested (long time) ([42, 40, 32, None], [None, 1, 5, 21])
- static UnitaryPolarGraph(m, q, algorithm='gap')#
Return the Unitary Polar Graph \(U(m,q)\).
For more information on Unitary Polar graphs, see the page of Andries Brouwer’s website.
INPUT:
m,q– integers; \(q\) must be a prime poweralgorithm– string (default:"gap"); if set to ‘gap’ then the computation is carried via GAP library interface, computing totally singular subspaces, which is faster for large examples (especially with \(q>2\)). Otherwise it is done directly.
EXAMPLES:
sage: # needs sage.libs.gap sage: G = graphs.UnitaryPolarGraph(4,2); G Unitary Polar Graph U(4, 2); GQ(4, 2): Graph on 45 vertices sage: G.is_strongly_regular(parameters=True) (45, 12, 3, 3) sage: graphs.UnitaryPolarGraph(5,2).is_strongly_regular(parameters=True) (165, 36, 3, 9) sage: graphs.UnitaryPolarGraph(6,2) # not tested (long time) Unitary Polar Graph U(6, 2): Graph on 693 vertices
- static UstimenkoGraph(m, q)#
Return the Ustimenko graph with parameters \((m, q)\).
This is the distance 1 or 2 graph of the dual polar graph \(C_{m-1}(q)\). The graph is distance-regular with classical with parameters \((d,q^2, qbinom(3,1,q) -1, qbinom(m+1,1,q) -1)\)
INPUT:
m, q– integers;qmust be a prime power andm > 1.
EXAMPLES:
sage: G = graphs.UstimenkoGraph(4, 2) # needs sage.libs.gap sage: G.is_distance_regular(True) # needs sage.libs.gap ([70, 32, None], [None, 1, 35])
REFERENCES:
See [BCN1989] p. 279 or [VDKT2016] p. 22.
- static WagnerGraph()#
Return the Wagner Graph.
See the Wikipedia article Wagner_graph.
EXAMPLES:
sage: # needs networkx sage: g = graphs.WagnerGraph() sage: g.order() 8 sage: g.size() 12 sage: g.girth() 4 sage: g.diameter() 2 sage: g.show() # needs sage.plot
- static WatkinsSnarkGraph()#
Return the Watkins Snark Graph.
The Watkins Graph is a snark with 50 vertices and 75 edges. For more information, see the Wikipedia article Watkins_snark.
EXAMPLES:
sage: g = graphs.WatkinsSnarkGraph() sage: g.order() 50 sage: g.size() 75 sage: g.chromatic_number() 3
- static WellsGraph()#
Return the Wells graph.
For more information on the Wells graph (also called Armanios-Wells graph), see this page.
The implementation follows the construction given on page 266 of [BCN1989]. This requires to create intermediate graphs and run a small isomorphism test, while everything could be replaced by a pre-computed list of edges. I believe that it is better to keep “the recipe” in the code, however, as it is quite unlikely that this could become the most time-consuming operation in any sensible algorithm, and …. “preserves knowledge”, which is what open-source software is meant to do.
EXAMPLES:
sage: g = graphs.WellsGraph(); g Wells graph: Graph on 32 vertices sage: g.order() 32 sage: g.size() 80 sage: g.girth() 5 sage: g.diameter() 4 sage: g.chromatic_number() 4 sage: g.is_regular(k=5) True
- static WheelGraph(n)#
Returns a Wheel graph with n nodes.
A Wheel graph is a basic structure where one node is connected to all other nodes and those (outer) nodes are connected cyclically.
PLOTTING: Upon construction, the position dictionary is filled to override the spring-layout algorithm. By convention, each wheel graph will be displayed with the first (0) node in the center, the second node at the top, and the rest following in a counterclockwise manner.
With the wheel graph, we see that it doesn’t take a very large n at all for the spring-layout to give a counter-intuitive display. (See Graphics Array examples below).
EXAMPLES:
We view many wheel graphs with a Sage Graphics Array, first with this constructor (i.e., the position dictionary filled):
sage: # needs sage.plot sage: g = [] sage: j = [] sage: for i in range(9): ....: k = graphs.WheelGraph(i+3) ....: g.append(k) ... sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) ... sage: G = graphics_array(j) sage: G.show() # long time
Next, using the spring-layout algorithm:
sage: # needs networkx sage.plot sage: import networkx sage: g = [] sage: j = [] sage: for i in range(9): ....: spr = networkx.wheel_graph(i+3) ....: k = Graph(spr) ....: g.append(k) ... sage: for i in range(3): ....: n = [] ....: for m in range(3): ....: n.append(g[3*i + m].plot(vertex_size=50, vertex_labels=False)) ....: j.append(n) ... sage: G = graphics_array(j) sage: G.show() # long time
Compare the plotting:
sage: # needs networkx sage.plot sage: n = networkx.wheel_graph(23) sage: spring23 = Graph(n) sage: posdict23 = graphs.WheelGraph(23) sage: spring23.show() # long time sage: posdict23.show() # long time
- static WienerArayaGraph()#
Return the Wiener-Araya Graph.
The Wiener-Araya Graph is a planar hypohamiltonian graph on 42 vertices and 67 edges. For more information, see the Wolfram Page on the Wiener-Araya Graph or Wikipedia article Wiener-Araya_graph.
EXAMPLES:
sage: g = graphs.WienerArayaGraph() sage: g.order() 42 sage: g.size() 67 sage: g.girth() 4 sage: g.is_planar() True sage: g.is_hamiltonian() # not tested (30s) # needs sage.numerical.mip False sage: g.delete_vertex(g.random_vertex()) sage: g.is_hamiltonian() # needs sage.numerical.mip True
- static WindmillGraph(k, n)#
Return the Windmill graph \(Wd(k, n)\).
The windmill graph \(Wd(k, n)\) is an undirected graph constructed for \(k \geq 2\) and \(n \geq 2\) by joining \(n\) copies of the complete graph \(K_k\) at a shared vertex. It has \((k-1)n+1\) vertices and \(nk(k-1)/2\) edges, girth 3 (if \(k > 2\)), radius 1 and diameter 2. It has vertex connectivity 1 because its central vertex is an articulation point; however, like the complete graphs from which it is formed, it is \((k-1)\)-edge-connected. It is trivially perfect and a block graph.
See also
EXAMPLES:
The Windmill graph \(Wd(2, n)\) is a star graph:
sage: n = 5 sage: W = graphs.WindmillGraph(2, n) sage: W.is_isomorphic( graphs.StarGraph(n) ) True
The Windmill graph \(Wd(3, n)\) is the Friendship graph \(F_n\):
sage: n = 5 sage: W = graphs.WindmillGraph(3, n) sage: W.is_isomorphic( graphs.FriendshipGraph(n) ) True
The Windmill graph \(Wd(3, 2)\) is the Butterfly graph:
sage: W = graphs.WindmillGraph(3, 2) sage: W.is_isomorphic( graphs.ButterflyGraph() ) True
The Windmill graph \(Wd(k, n)\) has chromatic number \(k\):
sage: n,k = 5,6 sage: W = graphs.WindmillGraph(k, n) sage: W.chromatic_number() == k True
- static WorldMap()#
Return the Graph of all the countries, in which two countries are adjacent in the graph if they have a common boundary.
This graph has been built from the data available in The CIA World Factbook [CIA] (2009-08-21).
The returned graph
Ghas a memberG.gps_coordinatesequal to a dictionary containing the GPS coordinates of each country’s capital city.EXAMPLES:
sage: g = graphs.WorldMap() sage: g.has_edge("France", "Italy") True sage: g.gps_coordinates["Bolivia"] [[17, 'S'], [65, 'W']] sage: g.connected_component_containing_vertex('Ireland', sort=True) ['Ireland', 'United Kingdom']
- basic = <module 'sage.graphs.generators.basic' from '/home/user/sage-git/src/sage/graphs/generators/basic.py'>#
- static chang_graphs()#
Return the three Chang graphs.
Three of the four strongly regular graphs of parameters \((28,12,6,4)\) are called the Chang graphs. The fourth is the line graph of \(K_8\). For more information about the Chang graphs, see the Wikipedia article Chang_graphs or https://www.win.tue.nl/~aeb/graphs/Chang.html.
EXAMPLES: check that we get 4 non-isomorphic s.r.g.’s with the same parameters:
sage: chang_graphs = graphs.chang_graphs() sage: K8 = graphs.CompleteGraph(8) sage: T8 = K8.line_graph() sage: four_srg = chang_graphs + [T8] sage: for g in four_srg: ....: print(g.is_strongly_regular(parameters=True)) (28, 12, 6, 4) (28, 12, 6, 4) (28, 12, 6, 4) (28, 12, 6, 4) sage: from itertools import combinations sage: for g1,g2 in combinations(four_srg,2): ....: assert not g1.is_isomorphic(g2)
Construct the Chang graphs by Seidel switching:
sage: c3c5 = graphs.CycleGraph(3).disjoint_union(graphs.CycleGraph(5)) sage: c8 = graphs.CycleGraph(8) sage: s = [K8.subgraph_search(c8).edges(sort=False), # needs sage.modules ....: [(0,1,None),(2,3,None),(4,5,None),(6,7,None)], ....: K8.subgraph_search(c3c5).edges(sort=False)] sage: [T8.seidel_switching(x, inplace=False).is_isomorphic(G) # needs sage.modules ....: for x, G in zip(s, chang_graphs)] [True, True, True]
- chessboard = <module 'sage.graphs.generators.chessboard' from '/home/user/sage-git/src/sage/graphs/generators/chessboard.py'>#
- classical_geometries = <module 'sage.graphs.generators.classical_geometries' from '/home/user/sage-git/src/sage/graphs/generators/classical_geometries.py'>#
- static cocliques_HoffmannSingleton()#
Return the graph obtained from the cocliques of the Hoffmann-Singleton graph.
This is a distance-regular graph with intersection array \([15, 14, 10, 3; 1, 5, 12, 15]\).
EXAMPLES:
sage: G = graphs.cocliques_HoffmannSingleton() sage: G.is_distance_regular(True) ([15, 14, 10, 3, None], [None, 1, 5, 12, 15])
REFERENCES:
The construction of this graph can be found in [BCN1989] p. 392.
- cospectral_graphs(vertices, matrix_function=None, graphs=None)#
Find all sets of graphs on
verticesvertices (with possible restrictions) which are cospectral with respect to a constructed matrix.INPUT:
vertices- The number of vertices in the graphs to be testedmatrix_function- A function taking a graph and giving back a matrix. This defaults to the adjacency matrix. The spectra examined are the spectra of these matrices.graphs- One of three things:None(default) - test all graphs havingverticesverticesa function taking a graph and returning
TrueorFalse- test only the graphs onverticesvertices for which the function returnsTruea list of graphs (or other iterable object) - these graphs are tested for cospectral sets. In this case,
verticesis ignored.
OUTPUT:
A list of lists of graphs. Each sublist will be a list of cospectral graphs (lists of cardinality 1 being omitted).
See also
Graph.is_strongly_regular()– tests whether a graph is strongly regular and/or returns its parameters.EXAMPLES:
sage: g = graphs.cospectral_graphs(5) # needs sage.modules sage: sorted(sorted(g.graph6_string() for g in glist) for glist in g) # needs sage.modules [['Dr?', 'Ds_']] sage: g[0][1].am().charpoly()==g[0][1].am().charpoly() # needs sage.modules True
There are two sets of cospectral graphs on six vertices with no isolated vertices:
sage: # needs sage.modules sage: g = graphs.cospectral_graphs(6, graphs=lambda x: min(x.degree())>0) sage: sorted(sorted(g.graph6_string() for g in glist) for glist in g) [['Ep__', 'Er?G'], ['ExGg', 'ExoG']] sage: g[0][1].am().charpoly()==g[0][1].am().charpoly() True sage: g[1][1].am().charpoly()==g[1][1].am().charpoly() True
There is one pair of cospectral trees on eight vertices:
sage: g = graphs.cospectral_graphs(6, graphs=graphs.trees(8)) # needs sage.modules sage: sorted(sorted(g.graph6_string() for g in glist) for glist in g) # needs sage.modules [['GiPC?C', 'GiQCC?']] sage: g[0][1].am().charpoly()==g[0][1].am().charpoly() # needs sage.modules True
There are two sets of cospectral graphs (with respect to the Laplacian matrix) on six vertices:
sage: # needs sage.modules sage: g = graphs.cospectral_graphs(6, matrix_function=lambda g: g.laplacian_matrix()) sage: sorted(sorted(g.graph6_string() for g in glist) for glist in g) [['Edq_', 'ErcG'], ['Exoo', 'EzcG']] sage: g[0][1].laplacian_matrix().charpoly()==g[0][1].laplacian_matrix().charpoly() True sage: g[1][1].laplacian_matrix().charpoly()==g[1][1].laplacian_matrix().charpoly() True
To find cospectral graphs with respect to the normalized Laplacian, assuming the graphs do not have an isolated vertex, it is enough to check the spectrum of the matrix \(D^{-1}A\), where \(D\) is the diagonal matrix of vertex degrees, and A is the adjacency matrix. We find two such cospectral graphs (for the normalized Laplacian) on five vertices:
sage: def DinverseA(g): ....: A = g.adjacency_matrix().change_ring(QQ) ....: for i in range(g.order()): ....: A.rescale_row(i, 1 / len(A.nonzero_positions_in_row(i))) ....: return A sage: g = graphs.cospectral_graphs(5, matrix_function=DinverseA, # needs sage.libs.pari sage.modules ....: graphs=lambda g: min(g.degree()) > 0) sage: sorted(sorted(g.graph6_string() for g in glist) for glist in g) # needs sage.modules [['Dlg', 'Ds_']] sage: (g[0][1].laplacian_matrix(normalized=True).charpoly() # needs sage.modules sage.symbolic ....: == g[0][1].laplacian_matrix(normalized=True).charpoly()) True
- degree_sequence = <module 'sage.graphs.generators.degree_sequence' from '/home/user/sage-git/src/sage/graphs/generators/degree_sequence.py'>#
- static distance_3_doubly_truncated_Golay_code_graph()#
Return a distance-regular graph with intersection array \([9, 8, 6, 3; 1, 1, 3, 8]\).
EXAMPLES:
sage: # long time, needs sage.modules sage.rings.finite_rings sage: G = graphs.distance_3_doubly_truncated_Golay_code_graph() sage: G.is_distance_regular(True) # long time (due to above) ([9, 8, 6, 3, None], [None, 1, 1, 3, 8])
ALGORITHM:
Compute the binary Golay code and truncate it twice. Compute its coset graph. Take a vertex and compute the set of vertices at distance 3 from the vertex chosen. This set constitutes the set of vertices of our distance-regular graph. Moreover we have an edge \((u,v)\) if the coset graph contains such edge.
REFERENCES:
Description and construction of this graph are taken from [BCN1989] p. 364.
- distance_regular = <module 'sage.graphs.generators.distance_regular' from '/home/user/sage-git/src/sage/graphs/generators/distance_regular.cpython-310-x86_64-linux-gnu.so'>#
- static distance_regular_graph(arr, existence=False, check=True)#
Return a distance-regular graph with the intersection array given.
INPUT:
arr– list; intersection array of the graphexistence– boolean (optional); instead of building the graph return:True- if a graph with the given intersection array exists;False- if there is no graph with the given intersection array;Unknown- if Sage doesn’t know if such a graph exists.
check– boolean (optional); ifTrue, then checks that the result of this function has the given intersection array. Default:True
EXAMPLES:
sage: graphs.distance_regular_graph([21,20,16,1,2,12], existence=True) True sage: G = graphs.distance_regular_graph([12,11,10,7,1,2,5,12], check=False) # needs sage.combinat sage.modules sage: G.is_distance_regular(True) # needs sage.combinat sage.modules ([12, 11, 10, 7, None], [None, 1, 2, 5, 12])
REFERENCES:
See [BCN1989] and [VDKT2016].
- families = <module 'sage.graphs.generators.families' from '/home/user/sage-git/src/sage/graphs/generators/families.py'>#
- fullerenes(order, ipr=False)#
Returns a generator which creates fullerene graphs using the buckygen generator (see [BGM2012]).
INPUT:
order- a positive even integer smaller than or equal to 254. This specifies the number of vertices in the generated fullerenes.ipr- default:False- ifTrueonly fullerenes that satisfy the Isolated Pentagon Rule are generated. This means that no pentagonal faces share an edge.
OUTPUT:
A generator which will produce the fullerene graphs as Sage graphs with an embedding set. These will be simple graphs: no loops, no multiple edges, no directed edges.
See also
set_embedding(),get_embedding()– get/set methods for embeddings.
EXAMPLES:
There are 1812 isomers of \(\textrm{C}_{60}\), i.e., 1812 fullerene graphs on 60 vertices:
sage: gen = graphs.fullerenes(60) # optional - buckygen sage: len(list(gen)) # optional - buckygen 1812
However, there is only one IPR fullerene graph on 60 vertices: the famous Buckminster Fullerene:
sage: gen = graphs.fullerenes(60, ipr=True) # optional - buckygen sage: next(gen) # optional - buckygen Graph on 60 vertices sage: next(gen) # optional - buckygen Traceback (most recent call last): ... StopIteration
The unique fullerene graph on 20 vertices is isomorphic to the dodecahedron graph.
sage: # optional - buckygen sage: gen = graphs.fullerenes(20) sage: g = next(gen) sage: g.is_isomorphic(graphs.DodecahedralGraph()) True sage: g.get_embedding() {1: [2, 3, 4], 2: [1, 5, 6], 3: [1, 7, 8], 4: [1, 9, 10], 5: [2, 10, 11], 6: [2, 12, 7], 7: [3, 6, 13], 8: [3, 14, 9], 9: [4, 8, 15], 10: [4, 16, 5], 11: [5, 17, 12], 12: [6, 11, 18], 13: [7, 18, 14], 14: [8, 13, 19], 15: [9, 19, 16], 16: [10, 15, 17], 17: [11, 16, 20], 18: [12, 20, 13], 19: [14, 20, 15], 20: [17, 19, 18]} sage: g.plot3d(layout='spring') Graphics3d Object
- fusenes(hexagon_count, benzenoids=False)#
Returns a generator which creates fusenes and benzenoids using the benzene generator (see [BCH2002]). Fusenes are planar polycyclic hydrocarbons with all bounded faces hexagons. Benzenoids are fusenes that are subgraphs of the hexagonal lattice.
INPUT:
hexagon_count- a positive integer smaller than or equal to 30. This specifies the number of hexagons in the generated benzenoids.benzenoids- default:False- ifTrueonly benzenoids are generated.
OUTPUT:
A generator which will produce the fusenes as Sage graphs with an embedding set. These will be simple graphs: no loops, no multiple edges, no directed edges.
See also
set_embedding(),get_embedding()– get/set methods for embeddings.
EXAMPLES:
There is a unique fusene with 2 hexagons:
sage: gen = graphs.fusenes(2) # optional - benzene sage: len(list(gen)) # optional - benzene 1
This fusene is naphthalene (\(\textrm{C}_{10}\textrm{H}_{8}\)). In the fusene graph the H-atoms are not stored, so this is a graph on just 10 vertices:
sage: gen = graphs.fusenes(2) # optional - benzene sage: next(gen) # optional - benzene Graph on 10 vertices sage: next(gen) # optional - benzene Traceback (most recent call last): ... StopIteration
There are 6505 benzenoids with 9 hexagons:
sage: gen = graphs.fusenes(9, benzenoids=True) # optional - benzene sage: len(list(gen)) # optional - benzene 6505
- static graph_3O73()#
Return the graph related to the group \(3 O(7,3)\).
This graph is distance-regular with intersection array \([117, 80, 24, 1; 1, 12, 80, 117]\).
The graph is also distance transitive with \(3.O(7,3)\) as automorphism group
EXAMPLES:
sage: G = graphs.graph_3O73() # optional - internet gap_package_atlasrep sage: G.is_distance_regular(True) # optional - internet gap_package_atlasrep ([117, 80, 24, 1, None], [None, 1, 12, 80, 117])
REFERENCES:
A description and construction of this graph can be found in [BCN1989] p. 400.
- intersection = <module 'sage.graphs.generators.intersection' from '/home/user/sage-git/src/sage/graphs/generators/intersection.py'>#
- static line_graph_forbidden_subgraphs()#
Returns the 9 forbidden subgraphs of a line graph.
See the Wikipedia article Line_graph for more information.
The graphs are returned in the ordering given by the Wikipedia drawing, read from left to right and from top to bottom.
EXAMPLES:
sage: graphs.line_graph_forbidden_subgraphs() [Claw graph: Graph on 4 vertices, Graph on 6 vertices, Graph on 6 vertices, Graph on 5 vertices, Graph on 6 vertices, Graph on 6 vertices, Graph on 6 vertices, Graph on 6 vertices, Graph on 5 vertices]
- static locally_GQ42_distance_transitive_graph()#
Return the unique amply regular graph with \(\mu = 6\) which is locally a generalised quadrangle.
This graph is distance-regular with intersection array \([45, 32, 12, 1; 1, 6, 32, 45]\).
This graph is also distance-transitive.
EXAMPLES:
sage: G = graphs.locally_GQ42_distance_transitive_graph() # optional - internet gap_package_atlasrep sage: G.is_distance_regular(True) # optional - internet gap_package_atlasrep ([45, 32, 12, 1, None], [None, 1, 6, 32, 45])
REFERENCES:
A description of this graph can be found in [BCN1989] p.399. This construction is due to Dima Pasechnik.
- nauty_genbg(options='', debug=False)#
Return a generator which creates bipartite graphs from nauty’s
genbgprogram.INPUT:
options– string (default:""); a string passed togenbgas if it was run at a system command line. At a minimum, you must pass the number of vertices you desire in each side. Sage expects the bipartite graphs to be in nauty’s “graph6” format, do not set an option to change this default or results will be unpredictable.debug– boolean (default:False); ifTruethe first line ofgeng’s output to standard error is captured and the first call to the generator’snext()function will return this line as a string. A line leading with “>A” indicates a successful initiation of the program with some information on the arguments, while a line beginning with “>E” indicates an error with the input.
The possible options, obtained as output of
genbg --help:n1 : the number of vertices in the first class. We must have n1=1..24. n2 : the number of vertices in the second class. We must have n2=0..32 and n1+n2=1..32. mine:maxe : <int>:<int> a range for the number of edges <int>:0 means '<int> or more' except in the case 0:0 res/mod : only generate subset res out of subsets 0..mod-1 -c : only write connected graphs -z : all the vertices in the second class must have different neighbourhoods -F : the vertices in the second class must have at least two neighbours of degree at least 2 -L : there is no vertex in the first class whose removal leaves the vertices in the second class unreachable from each other -Y<int> : two vertices in the second class must have at least <int> common neighbours -Z<int> : two vertices in the second class must have at most <int> common neighbours -A : no vertex in the second class has a neighbourhood which is a subset of another vertex's neighbourhood in the second class -D<int> : specify an upper bound for the maximum degree Example: -D6. You can also give separate maxima for the two parts, for example: -D5:6 -d<int> : specify a lower bound for the minimum degree Again, you can specify it separately for the two parts, for example -d1:2 -v : display counts by number of edges to stderr -l : canonically label output graphs
Options which cause
genbgto use an output format different than thegraph6format are not listed above (-s,-a) as they will confuse the creation of a Sage graph. Option-qwhich suppress auxiliary output (except from-v) should never be used as we are unable to recover the partition of the vertices of the bipartite graph without the auxiliary output. Hence the partition of the vertices of returned bipartite graphs might not respect the requirement.The res/mod option can be useful when using the output in a routine run several times in parallel.
OUTPUT:
A generator which will produce the graphs as
BipartiteGraph. These will be simple bipartite graphs: no loops, no multiple edges, no directed edges.EXAMPLES:
The generator can be used to construct biparrtite graphs for testing, one at a time (usually inside a loop). Or it can be used to create an entire list all at once if there is sufficient memory to contain it:
sage: gen = graphs.nauty_genbg("1 1") sage: next(gen) Bipartite graph on 2 vertices sage: next(gen) Bipartite graph on 2 vertices sage: next(gen) Traceback (most recent call last): ... StopIteration
Connected bipartite graphs of order 6 with different number of vertices in each side:
sage: gen = graphs.nauty_genbg("1 5 -c") sage: len(list(gen)) 1 sage: gen = graphs.nauty_genbg("2 4 -c") sage: len(list(gen)) 6 sage: gen = graphs.nauty_genbg("3 3 -c") sage: len(list(gen)) 13
Use
nauty_geng()instead if you want the list of all bipartite graphs of order \(n\). For instance, the list of all connected bipartite graphs of order 6, which agrees with OEIS sequence A005142:sage: gen = graphs.nauty_geng("-b -c 6") sage: len(list(gen)) 17
The
debugswitch can be used to examinegenbg’s reaction to the input in theoptionsstring. A message starting with “>A” indicates success and a message starting with “>E” indicates a failure:sage: gen = graphs.nauty_genbg("2 3", debug=True) sage: print(next(gen)) >A ...genbg n=2+3 e=0:6 d=0:0 D=3:2 sage: gen = graphs.nauty_genbg("-c2 3", debug=True) sage: next(gen) '>E Usage: ...genbg [-c -ugs -vq -lzF] [-Z#] [-D#] [-A] [-d#|-d#:#] [-D#|-D#:#] n1 n2...
Check that the partition of the bipartite graph is consistent:
sage: gen = graphs.nauty_genbg("3 3") sage: left = set(range(3)) sage: for g in gen: ....: if g.left != left: ....: raise ValueError('wrong partition')
- nauty_geng(options='', debug=False)#
Return a generator which creates graphs from nauty’s geng program.
INPUT:
options– string (default:""); a string passed togengas if it was run at a system command line. At a minimum, you must pass the number of vertices you desire. Sage expects the graphs to be in nauty’s “graph6” format, do not set an option to change this default or results will be unpredictable.debug– boolean (default:False); ifTruethe first line ofgeng’s output to standard error is captured and the first call to the generator’snext()function will return this line as a string. A line leading with “>A” indicates a successful initiation of the program with some information on the arguments, while a line beginning with “>E” indicates an error with the input.
The possible options, obtained as output of
geng --help:n : the number of vertices mine:maxe : <int>:<int> a range for the number of edges <int>:0 means '<int> or more' except in the case 0:0 res/mod : only generate subset res out of subsets 0..mod-1 -c : only write connected graphs -C : only write biconnected graphs -t : only generate triangle-free graphs -f : only generate 4-cycle-free graphs -b : only generate bipartite graphs (-t, -f and -b can be used in any combination) -m : save memory at the expense of time (only makes a difference in the absence of -b, -t, -f and n <= 28). -d<int> : a lower bound for the minimum degree -D<int> : a upper bound for the maximum degree -v : display counts by number of edges -l : canonically label output graphs -q : suppress auxiliary output (except from -v)
Options which cause
gengto use an output format different than the graph6 format are not listed above (-u, -g, -s, -y, -h) as they will confuse the creation of a Sage graph. The res/mod option can be useful when using the output in a routine run several times in parallel.OUTPUT:
A generator which will produce the graphs as Sage graphs. These will be simple graphs: no loops, no multiple edges, no directed edges.
See also
Graph.is_strongly_regular()– tests whether a graph is strongly regular and/or returns its parameters.EXAMPLES:
The generator can be used to construct graphs for testing, one at a time (usually inside a loop). Or it can be used to create an entire list all at once if there is sufficient memory to contain it.
sage: gen = graphs.nauty_geng("2") sage: next(gen) Graph on 2 vertices sage: next(gen) Graph on 2 vertices sage: next(gen) Traceback (most recent call last): ... StopIteration
A list of all graphs on 7 vertices. This agrees with OEIS sequence A000088.
sage: gen = graphs.nauty_geng("7") sage: len(list(gen)) 1044
A list of just the connected graphs on 7 vertices. This agrees with OEIS sequence A001349.
sage: gen = graphs.nauty_geng("7 -c") sage: len(list(gen)) 853
A list of connected degree exactly 2 graphs on 5 vertices.
sage: gen = graphs.nauty_geng("5 -c -d2 -D2") sage: len(list(gen)) 1
The
debugswitch can be used to examinegeng’s reaction to the input in theoptionsstring. We illustrate success. (A failure will be a string beginning with “>E”.) Passing the “-q” switch togengwill suppress the indicator of a successful initiation, and so the first returned value might be an empty string ifdebugisTrue:sage: gen = graphs.nauty_geng("4", debug=True) sage: print(next(gen)) >A ...geng -d0D3 n=4 e=0-6 sage: gen = graphs.nauty_geng("4 -q", debug=True) sage: next(gen) ''
- nauty_genktreeg(options='', debug=False)#
Return a generator which creates all \(k\)-trees using nauty..
A \(k\)-tree is an undirected graph formed by starting with a complete graph on \(k + 1\) vertices and then repeatedly add vertices in such a way that each added vertex \(v\) has exactly \(k\) neighbors \(U\) such that, together, the \(k + 1\) vertices formed by \(v\) and \(U\) form a clique. See the Wikipedia article K-tree for more details.
INPUT:
options– string (default:""); a string passed togenktreegas if it was run at a system command line. At a minimum, you must pass the number of vertices you desire. Sage expects the graphs to be in nauty’s “graph6” format, do not set an option to change this default or results will be unpredictable.debug– boolean (default:False); ifTruethe first line ofgenktreeg’s output to standard error is captured and the first call to the generator’snext()function will return this line as a string. A line leading with “>A” indicates a successful initiation of the program with some information on the arguments, while a line beginning with “>E” indicates an error with the input.
The possible options, obtained as output of
genktreeg --help:n : the number of vertices -k<int> : the value of `k`(default: 2) res/mod : only generate subset res out of subsets 0..mod-1 -l : canonically label output graphs
Options which cause
genktreegto use an output format different than the graph6 format are not listed above (-u, -s, -h) as they will confuse the creation of a Sage graph. The res/mod option can be useful when using the output in a routine run several times in parallel.OUTPUT:
A generator which will produce the graphs as Sage graphs. These will be simple graphs: no loops, no multiple edges, no directed edges.
EXAMPLES:
A \(k\)-tree is a maximal graph with treewidth \(k\):
sage: # needs nauty sage: gen = graphs.nauty_genktreeg("10 -k4") sage: G = next(gen); G Graph on 10 vertices sage: G.treewidth() 4
A list of all 2-trees with 6, 7 and 8 vertices. This agrees with OEIS sequence A054581:
sage: # needs nauty sage: gen = graphs.nauty_genktreeg("6") sage: len(list(gen)) 5 sage: gen = graphs.nauty_genktreeg("7") sage: len(list(gen)) 12 sage: gen = graphs.nauty_genktreeg("8") sage: len(list(gen)) 39
The
debugswitch can be used to examinegeng’s reaction to the input in theoptionsstring. We illustrate success. (A failure will be a string beginning with “>E”.) Passing the “-q” switch togengwill suppress the indicator of a successful initiation, and so the first returned value might be an empty string ifdebugisTrue:sage: gen = graphs.nauty_genktreeg("7", debug=True) # needs nauty sage: print(next(gen)) # needs nauty >A ...genktreeg k=2 n=7
- static nauty_gentreeg(options='', debug=False)#
Return a generator which creates non-isomorphic trees from nauty’s gentreeg program.
INPUT:
options– string (default:""); a string passed togentreegas if it was run at a system command line. At a minimum, you must pass the number of vertices you desire. Sage expects the graphs to be in nauty’s “sparse6” format, do not set an option to change this default or results will be unpredictable.debug– boolean (default:False); ifTruethe first line ofgentreeg’s output to standard error is captured and the first call to the generator’snext()function will return this line as a string. A line leading with “>A” indicates a successful initiation of the program with some information on the arguments, while a line beginning with “>E” indicates an error with the input.
The possible options, obtained as output of
gentreeg -help:n : the number of vertices. Must be in range 1..128 res/mod : only generate subset res out of subsets 0..mod-1 -D<int> : an upper bound for the maximum degree -Z<int>:<int> : bounds on the diameter -q : suppress auxiliary output
Options which cause
gentreegto use an output format different than the sparse6 format are not listed above (-p, -l, -u) as they will confuse the creation of a Sage graph. The res/mod option can be useful when using the output in a routine run several times in parallel.OUTPUT:
A generator which will produce the graphs as Sage graphs. These will be simple graphs: no loops, no multiple edges, no directed edges.
See also
trees()– another generator of treesEXAMPLES:
The generator can be used to construct trees for testing, one at a time (usually inside a loop). Or it can be used to create an entire list all at once if there is sufficient memory to contain it:
sage: gen = graphs.nauty_gentreeg("4") sage: next(gen) Graph on 4 vertices sage: next(gen) Graph on 4 vertices sage: next(gen) Traceback (most recent call last): ... StopIteration
The number of trees on the first few vertex counts. This agrees with OEIS sequence A000055:
sage: [len(list(graphs.nauty_gentreeg(str(i)))) for i in range(1, 15)] [1, 1, 1, 2, 3, 6, 11, 23, 47, 106, 235, 551, 1301, 3159]
The
debugswitch can be used to examinegentreeg’s reaction to the input in theoptionsstring. We illustrate success. (A failure will be a string beginning with “>E”.) Passing the “-q” switch togentreegwill suppress the indicator of a successful initiation, and so the first returned value might be an empty string ifdebugisTrue:sage: gen = graphs.nauty_gentreeg("4", debug=True) sage: print(next(gen)) >A ...gentreeg ... sage: gen = graphs.nauty_gentreeg("4 -q", debug=True) sage: next(gen) ''
- static petersen_family(generate=False)#
Returns the Petersen family
The Petersen family is a collection of 7 graphs which are the forbidden minors of the linklessly embeddable graphs. For more information see the Wikipedia article Petersen_family.
INPUT:
generate(boolean) – whether to generate the family from the \(\Delta-Y\) transformations. When set toFalse(default) a hardcoded version of the graphs (with a prettier layout) is returned.
EXAMPLES:
sage: graphs.petersen_family() [Petersen graph: Graph on 10 vertices, Complete graph: Graph on 6 vertices, Multipartite Graph with set sizes [3, 3, 1]: Graph on 7 vertices, Graph on 8 vertices, Graph on 9 vertices, Graph on 7 vertices, Graph on 8 vertices]
The two different inputs generate the same graphs:
sage: F1 = graphs.petersen_family(generate=False) sage: F2 = graphs.petersen_family(generate=True) # needs sage.modules sage: F1 = [g.canonical_label().graph6_string() for g in F1] sage: F2 = [g.canonical_label().graph6_string() for g in F2] # needs sage.modules sage: set(F1) == set(F2) # needs sage.modules True
- planar_graphs(order, minimum_degree=None, minimum_connectivity=None, exact_connectivity=False, minimum_edges=None, maximum_edges=None, maximum_face_size=None, only_bipartite=False, dual=False)#
An iterator over connected planar graphs using the plantri generator.
This uses the plantri generator (see [BM2007]) which is available through the optional package plantri.
Note
The non-3-connected graphs will be returned several times, with all its possible embeddings.
INPUT:
order- a positive integer smaller than or equal to 64. This specifies the number of vertices in the generated graphs.minimum_degree- default:None- a value \(\geq 1\) and \(\leq 5\), orNone. This specifies the minimum degree of the generated graphs. If this isNoneand the order is 1, then this is set to 0. If this isNoneand the minimum connectivity is specified, then this is set to the same value as the minimum connectivity. If the minimum connectivity is also equal toNone, then this is set to 1.minimum_connectivity- default:None- a value \(\geq 1\) and \(\leq 3\), orNone. This specifies the minimum connectivity of the generated graphs. If this isNoneand the minimum degree is specified, then this is set to the minimum of the minimum degree and 3. If the minimum degree is also equal toNone, then this is set to 1.exact_connectivity- default:False- ifTrueonly graphs with exactly the specified connectivity will be generated. This option cannot be used withminimum_connectivity=3, or if the minimum connectivity is not explicitly set.minimum_edges– integer (default:None); lower bound on the number of edgesmaximum_edges– integer (default:None); upper bound on the number of edgesmaximum_face_size– integer (default:None); upper bound on the size of a face and so on the maximum degree of the dual graphonly_bipartite- default:False- ifTrueonly bipartite graphs will be generated. This option cannot be used for graphs with a minimum degree larger than 3.dual- default:False- ifTruereturn instead the planar duals of the generated graphs.
OUTPUT:
An iterator which will produce all planar graphs with the given number of vertices as Sage graphs with an embedding set. These will be simple graphs (no loops, no multiple edges, no directed edges) unless the option
dual=Trueis used.See also
set_embedding(),get_embedding()– get/set methods for embeddings.
EXAMPLES:
There are 6 planar graphs on 4 vertices:
sage: gen = graphs.planar_graphs(4) # optional - plantri sage: len(list(gen)) # optional - plantri 6
Three of these planar graphs are bipartite:
sage: gen = graphs.planar_graphs(4, only_bipartite=True) # optional - plantri sage: len(list(gen)) # optional - plantri 3
Setting
dual=Truegives the planar dual graphs:sage: gen = graphs.planar_graphs(4, dual=True) # optional - plantri sage: [u for u in list(gen)] # optional - plantri [Graph on 4 vertices, Multi-graph on 3 vertices, Multi-graph on 2 vertices, Looped multi-graph on 2 vertices, Looped multi-graph on 1 vertex, Looped multi-graph on 1 vertex]
The cycle of length 4 is the only 2-connected bipartite planar graph on 4 vertices:
sage: l = list(graphs.planar_graphs(4, minimum_connectivity=2, only_bipartite=True)) # optional - plantri sage: l[0].get_embedding() # optional - plantri {1: [2, 3], 2: [1, 4], 3: [1, 4], 4: [2, 3]}
There is one planar graph with one vertex. This graph obviously has minimum degree equal to 0:
sage: list(graphs.planar_graphs(1)) # optional - plantri [Graph on 1 vertex] sage: list(graphs.planar_graphs(1, minimum_degree=1)) # optional - plantri []
Specifying lower and upper bounds on the number of edges:
sage: # optional - plantri sage: len(list(graphs.planar_graphs(4))) 6 sage: len(list(graphs.planar_graphs(4, minimum_edges=4))) 4 sage: len(list(graphs.planar_graphs(4, maximum_edges=4))) 4 sage: len(list(graphs.planar_graphs(4, minimum_edges=4, maximum_edges=4))) 2
Specifying the maximum size of a face:
sage: len(list(graphs.planar_graphs(4, maximum_face_size=3))) # optional - plantri 1 sage: len(list(graphs.planar_graphs(4, maximum_face_size=4))) # optional - plantri 3
- plantri_gen(options='')#
Iterator over planar graphs created using the
plantrigenerator.plantriis a (optional) program that generates certain types of graphs that are embedded on the sphere. It outputs exactly one member of each isomorphism class, using an amount of memory almost independent of the number of graphs produced. Isomorphisms are defined with respect to the embeddings, so in some cases outputs may be isomorphic as abstract graphs.This method allows for passing command directly to
plantry, similarly to methodnauty_geng(), provide that the output format is not changed.INPUT:
options– string (default:""); a string passed toplantrias if it was run at a system command line. At a minimum, you must pass the number of vertices you desire. Sage expects the output of plantri to be in “planar code” format, so do not set an option to change this default or results will be unpredictable.
The possible options are:
n : the number of vertices (the only compulsory parameter). This number must be in range `3\cdots 64`. It can also be given as "nd", where the suffix "d" means "dual", in which case it is converted by adding 4 then dividing by 2, i.e., `(28+4)/2 = 16`. In the case of triangulations, this calculation yields the number of faces, which is the number of vertices in the dual cubic graph. -d : output the dual instead of the original graph. Note that it is applied only at the output stage. All other switches refer to the original graph before the dual is taken. -o : Normally, one member of each isomorphism class is written. If this switch is given, one member of each O-P isomorphism class is written. -V : output only graphs with non-trivial group. If -o is given the O-P group is used, the full group otherwise. -m<int> : lower bound on the minimum degree. The default is -m3. In the dual graph, this means a lower bound on the minimum face size. -c<int> : lower bound on the connectivity. The default is -c3. -x : when used in combination with -cN, the connectivity must be exactly N rather than at least N. -e : used to specify bounds on the number of edges. There are four possible forms: -e<int> exactly <int> edges -e:<int> at most <int> edges -e<int>: at least <int> edges -e<int>:<int> between <int> and <int> edges -f<int> : upper bound on the size of a face, and so on the maximum degree of the dual. -b but not -p : select eulerian triangulations, where "eulerian" means that every vertex has even degree. This parameter can be used in combination with parameters -c and -x. -p but not -b : select general planar simple graphs. This parameter can be used in combination with parameters -m, -c, -x, -e and -f. -bp or -pb : select general planar simple bipartite graphs. This parameter can be used in combination with parameters -m, -c, -x, -e and -f, except -c4, -m4, -m5 and -f3. -P<int> : select triangulations of a disk. These are embedded simple graphs with a distinguished "outer" face. The outer face can be of any size (here called the disk size) but the other faces must be triangles. The argument <int> to -P is the disk size. If no argument (or 0) is given, all disk sizes are permitted. This parameter can be used in combination with parameters -m, -c, and -x. -q : select simple quadrangulations. These are planar simple graphs for which every face has length 4. This parameter can be used in combination with parameters -c and -m. -A : select Appolonian networks. These are simple planar triangulations that can be formed starting with `K_4` then repeatedly dividing a face into three by addition of a new vertex. They all have minimum degree and connectivity equal to 3. res/mod : only generate subset res out of subsets 0..mod-1. The set of objects is divided into mod disjoint classes and only the res-th class is generated.
If -b, -q, -p, -P and -A are absent, the graphs found are triangulations only restricted by connectivity and minimum degree. In this case, there is the possibility of connectivity lower than 3.
Other options listed in the
plantriguide might cause unpredictable behavior, in particular those changing the output format ofplantrias they will confuse the creation of a Sage graph.OUTPUT:
An iterator which yields the graphs generated by
plantrias SageGraph.See also
planar_graphs()– iterator over connected planar graphs using theplantrigeneratortriangulations()– iterator over connected planar triangulations using theplantrigeneratorquadrangulations()– iterator over connected planar quadrangulations using theplantrigenerator
EXAMPLES:
The generator can be used to construct graphs for testing, one at a time (usually inside a loop). Or it can be used to create an entire list all at once if there is sufficient memory to contain it:
sage: # optional - plantri sage: gen = graphs.plantri_gen("6") sage: next(gen) Graph on 6 vertices sage: next(gen) Graph on 6 vertices sage: next(gen) Traceback (most recent call last): ... StopIteration
An overview of the number of quadrangulations on up to 12 vertices. This agrees with OEIS sequence A113201:
sage: for i in range(4, 13): # optional - plantri ....: cmd = '-qm2c2 {}'.format(i) ....: L = len(list(graphs.plantri_gen(cmd))) ....: print("{:2d} {:3d}".format(i, L)) 4 1 5 1 6 2 7 3 8 9 9 18 10 62 11 198 12 803
- platonic_solids = <module 'sage.graphs.generators.platonic_solids' from '/home/user/sage-git/src/sage/graphs/generators/platonic_solids.py'>#
- quadrangulations(order, minimum_degree=None, minimum_connectivity=None, no_nonfacial_quadrangles=False, dual=False)#
An iterator over planar quadrangulations using the plantri generator.
This uses the plantri generator (see [BM2007]) which is available through the optional package plantri.
INPUT:
order- a positive integer smaller than or equal to 64. This specifies the number of vertices in the generated quadrangulations.minimum_degree- default:None- a value \(\geq 2\) and \(\leq 3\), orNone. This specifies the minimum degree of the generated quadrangulations. If this isNoneand the minimum connectivity is specified, then this is set to the same value as the minimum connectivity. If the minimum connectivity is also equal toNone, then this is set to 2.minimum_connectivity- default:None- a value \(\geq 2\) and \(\leq 3\), orNone. This specifies the minimum connectivity of the generated quadrangulations. If this isNoneand the optionno_nonfacial_quadranglesis set toTrue, then this is set to 3. Otherwise if this isNoneand the minimum degree is specified, then this is set to the minimum degree. If the minimum degree is also equal toNone, then this is set to 3.no_nonfacial_quadrangles- default:False- ifTrueonly quadrangulations with no non-facial quadrangles are generated. This option cannot be used ifminimum_connectivityis set to 2.dual- default:False- ifTruereturn instead the planar duals of the generated graphs.
OUTPUT:
An iterator which will produce all planar quadrangulations with the given number of vertices as Sage graphs with an embedding set. These will be simple graphs (no loops, no multiple edges, no directed edges).
See also
set_embedding(),get_embedding()– get/set methods for embeddings.
EXAMPLES:
The cube is the only 3-connected planar quadrangulation on 8 vertices:
sage: # optional - plantri sage: gen = graphs.quadrangulations(8, minimum_connectivity=3) sage: g = next(gen) sage: g.is_isomorphic(graphs.CubeGraph(3)) True sage: next(gen) Traceback (most recent call last): ... StopIteration
An overview of the number of quadrangulations on up to 12 vertices. This agrees with OEIS sequence A113201:
sage: for i in range(4,13): # optional - plantri ....: L = len(list(graphs.quadrangulations(i))) ....: print("{:2d} {:3d}".format(i,L)) 4 1 5 1 6 2 7 3 8 9 9 18 10 62 11 198 12 803
There are 2 planar quadrangulation on 12 vertices that do not have a non-facial quadrangle:
sage: len([g for g in graphs.quadrangulations(12, no_nonfacial_quadrangles=True)]) # optional - plantri 2
Setting
dual=Truegives the planar dual graphs:sage: [len(g) for g in graphs.quadrangulations(12, no_nonfacial_quadrangles=True, dual=True)] # optional - plantri [10, 10]
- random = <module 'sage.graphs.generators.random' from '/home/user/sage-git/src/sage/graphs/generators/random.py'>#
- static shortened_000_111_extended_binary_Golay_code_graph()#
Return a distance-regular graph with intersection array \([21, 20, 16, 9, 2, 1; 1, 2, 3, 16, 20, 21]\).
EXAMPLES:
sage: # long time, needs sage.modules sage.rings.finite_rings sage: G = graphs.shortened_000_111_extended_binary_Golay_code_graph() # 25 s sage: G.is_distance_regular(True) ([21, 20, 16, 9, 2, 1, None], [None, 1, 2, 3, 16, 20, 21])
ALGORITHM:
Compute the extended binary Golay code. Compute its subcode whose codewords start with 000 or 111. Remove the first 3 entries from all the codewords from the new linear code and compute its coset graph.
REFERENCES:
Description and construction of this graph can be found in [BCN1989] p. 365.
- static shortened_00_11_binary_Golay_code_graph()#
Return a distance-regular graph with intersection array \([21, 20, 16, 6, 2, 1; 1, 2, 6, 16, 20, 21]\).
EXAMPLES:
sage: # long time, needs sage.modules sage.rings.finite_rings sage: G = graphs.shortened_00_11_binary_Golay_code_graph() # 9 s sage: G.is_distance_regular(True) ([21, 20, 16, 6, 2, 1, None], [None, 1, 2, 6, 16, 20, 21])
ALGORITHM:
Compute the binary Golay code. Compute the subcode whose codewords start with 00 or 11. Remove the first two entries from all codewords of the newly found linear code and compute its coset graph.
REFERENCES:
Description and construction of this graph can be found in [BCN1989] p. 365.
- smallgraphs = <module 'sage.graphs.generators.smallgraphs' from '/home/user/sage-git/src/sage/graphs/generators/smallgraphs.py'>#
- strongly_regular_db = <module 'sage.graphs.strongly_regular_db' from '/home/user/sage-git/src/sage/graphs/strongly_regular_db.cpython-310-x86_64-linux-gnu.so'>#
- static strongly_regular_graph(v, k, l, mu=-1, existence=False, check=True)#
Return a \((v,k,\lambda,\mu)\)-strongly regular graph.
This function relies partly on Andries Brouwer’s database of strongly regular graphs. See the documentation of
sage.graphs.strongly_regular_dbfor more information.INPUT:
v,k,l,mu(integers) – note thatmu, if unspecified, is automatically determined fromv,k,l.existence(boolean;``False``) – instead of building the graph, return:True– meaning that a \((v,k,\lambda,\mu)\)-strongly regular graph exists.Unknown– meaning that Sage does not know if such a strongly regular graph exists (seesage.misc.unknown).False– meaning that no such strongly regular graph exists.
check– (boolean) Whether to check that output is correct before returning it. As this is expected to be useless (but we are cautious guys), you may want to disable it whenever you want speed. Set toTrueby default.
EXAMPLES:
Petersen’s graph from its set of parameters:
sage: graphs.strongly_regular_graph(10,3,0,1,existence=True) # needs sage.libs.pari True sage: graphs.strongly_regular_graph(10,3,0,1) complement(Johnson graph with parameters 5,2): Graph on 10 vertices
Now without specifying \(\mu\):
sage: graphs.strongly_regular_graph(10,3,0) complement(Johnson graph with parameters 5,2): Graph on 10 vertices
An obviously infeasible set of parameters:
sage: graphs.strongly_regular_graph(5,5,5,5,existence=True) False sage: graphs.strongly_regular_graph(5,5,5,5) Traceback (most recent call last): ... ValueError: There exists no (5, 5, 5, 5)-strongly regular graph
A set of parameters proved in a paper to be infeasible:
sage: graphs.strongly_regular_graph(324,57,0,12,existence=True) # needs sage.combinat sage.modules False sage: graphs.strongly_regular_graph(324,57,0,12) # needs sage.combinat sage.modules Traceback (most recent call last): ... EmptySetError: Andries Brouwer's database reports that no (324, 57, 0, 12)-strongly regular graph exists. Comments: <a href="srgtabrefs.html#GavrilyukMakhnev05">Gavrilyuk & Makhnev</a> ...
A set of parameters unknown to be realizable in Andries Brouwer’s database:
sage: graphs.strongly_regular_graph(324,95,22,30,existence=True) # needs sage.combinat Unknown sage: graphs.strongly_regular_graph(324,95,22,30) # needs sage.combinat Traceback (most recent call last): ... RuntimeError: Andries Brouwer's database reports that no (324, 95, 22, 30)-strongly regular graph is known to exist. Comments:
A large unknown set of parameters (not in Andries Brouwer’s database):
sage: graphs.strongly_regular_graph(1394,175,0,25,existence=True) # needs sage.combinat Unknown sage: graphs.strongly_regular_graph(1394,175,0,25) # needs sage.combinat Traceback (most recent call last): ... RuntimeError: Sage cannot figure out if a (1394, 175, 0, 25)-strongly regular graph exists.
Test the Claw bound (see 3.D of [BL1984]):
sage: graphs.strongly_regular_graph(2058,242,91,20,existence=True) False
- static trees(vertices)#
Returns a generator of the distinct trees on a fixed number of vertices.
INPUT:
vertices- the size of the trees created.
OUTPUT:
A generator which creates an exhaustive, duplicate-free listing of the connected free (unlabeled) trees with
verticesnumber of vertices. A tree is a graph with no cycles.ALGORITHM:
Uses an algorithm that generates each new tree in constant time. See the documentation for, and implementation of, the
sage.graphs.treesmodule, including a citation.EXAMPLES:
We create an iterator, then loop over its elements.
sage: tree_iterator = graphs.trees(7) sage: for T in tree_iterator: ....: print(T.degree_sequence()) [2, 2, 2, 2, 2, 1, 1] [3, 2, 2, 2, 1, 1, 1] [3, 2, 2, 2, 1, 1, 1] [4, 2, 2, 1, 1, 1, 1] [3, 3, 2, 1, 1, 1, 1] [3, 3, 2, 1, 1, 1, 1] [4, 3, 1, 1, 1, 1, 1] [3, 2, 2, 2, 1, 1, 1] [4, 2, 2, 1, 1, 1, 1] [5, 2, 1, 1, 1, 1, 1] [6, 1, 1, 1, 1, 1, 1]
The number of trees on the first few vertex counts. This is sequence A000055 in Sloane’s OEIS.
sage: [len(list(graphs.trees(i))) for i in range(0, 15)] [1, 1, 1, 1, 2, 3, 6, 11, 23, 47, 106, 235, 551, 1301, 3159]
- triangulations(order, minimum_degree=None, minimum_connectivity=None, exact_connectivity=False, only_eulerian=False, dual=False)#
An iterator over connected planar triangulations using the plantri generator.
This uses the plantri generator (see [BM2007]) which is available through the optional package plantri.
INPUT:
order- a positive integer smaller than or equal to 64. This specifies the number of vertices in the generated triangulations.minimum_degree- default:None- a value \(\geq 3\) and \(\leq 5\), orNone. This specifies the minimum degree of the generated triangulations. If this isNoneand the minimum connectivity is specified, then this is set to the same value as the minimum connectivity. If the minimum connectivity is also equal toNone, then this is set to 3.minimum_connectivity- default:None- a value \(\geq 3\) and \(\leq 5\), orNone. This specifies the minimum connectivity of the generated triangulations. If this isNoneand the minimum degree is specified, then this is set to the minimum of the minimum degree and 3. If the minimum degree is also equal toNone, then this is set to 3.exact_connectivity- default:False- ifTrueonly triangulations with exactly the specified connectivity will be generated. This option cannot be used withminimum_connectivity=3, or if the minimum connectivity is not explicitly set.only_eulerian- default:False- ifTrueonly Eulerian triangulations will be generated. This option cannot be used if the minimum degree is explicitly set to anything else than 4.dual- default:False- ifTruereturn instead the planar duals of the generated graphs.
OUTPUT:
An iterator which will produce all planar triangulations with the given number of vertices as Sage graphs with an embedding set. These will be simple graphs (no loops, no multiple edges, no directed edges).
See also
set_embedding(),get_embedding()– get/set methods for embeddings.RandomTriangulation()– build a random triangulation.
EXAMPLES:
The unique planar embedding of the \(K_4\) is the only planar triangulations on 4 vertices:
sage: gen = graphs.triangulations(4) # optional - plantri sage: [g.get_embedding() for g in gen] # optional - plantri [{1: [2, 3, 4], 2: [1, 4, 3], 3: [1, 2, 4], 4: [1, 3, 2]}]
but, of course, this graph is not Eulerian:
sage: gen = graphs.triangulations(4, only_eulerian=True) # optional - plantri sage: len(list(gen)) # optional - plantri 0
The unique Eulerian triangulation on 6 vertices is isomorphic to the octahedral graph.
sage: gen = graphs.triangulations(6, only_eulerian=True) # optional - plantri sage: g = next(gen) # optional - plantri sage: g.is_isomorphic(graphs.OctahedralGraph()) # optional - plantri True
An overview of the number of 5-connected triangulations on up to 22 vertices. This agrees with OEIS sequence A081621:
sage: for i in range(12, 23): # optional - plantri ....: L = len(list(graphs.triangulations(i, minimum_connectivity=5))) ....: print("{} {:3d}".format(i,L)) 12 1 13 0 14 1 15 1 16 3 17 4 18 12 19 23 20 71 21 187 22 627
The minimum connectivity can be at most the minimum degree:
sage: gen = next(graphs.triangulations(10, minimum_degree=3, # optional - plantri ....: minimum_connectivity=5)) Traceback (most recent call last): ... ValueError: Minimum connectivity can be at most the minimum degree.
There are 5 triangulations with 9 vertices and minimum degree equal to 4 that are 3-connected, but only one of them is not 4-connected:
sage: len([g for g in graphs.triangulations(9, minimum_degree=4, # optional - plantri ....: minimum_connectivity=3)]) 5 sage: len([g for g in graphs.triangulations(9, minimum_degree=4, # optional - plantri ....: minimum_connectivity=3, ....: exact_connectivity=True)]) 1
Setting
dual=Truegives the planar dual graphs:sage: [len(g) for g in graphs.triangulations(9, minimum_degree=4, # optional plantri ....: minimum_connectivity=3, dual=True)] [14, 14, 14, 14, 14]
- static vanLintSchrijverGraph()#
Return the van Lint-Schrijver graph.
The graph is distance-regular with intersection array \([6, 5, 5, 4; 1, 1, 2, 6]\).
EXAMPLES:
sage: G = graphs.vanLintSchrijverGraph() # needs sage.modules sage: G.is_distance_regular(True) # needs sage.modules ([6, 5, 5, 4, None], [None, 1, 1, 2, 6])
REFERENCES:
For a description of this graph see [BCN1989] p. 373.
- world_map = <module 'sage.graphs.generators.world_map' from '/home/user/sage-git/src/sage/graphs/generators/world_map.py'>#
- sage.graphs.graph_generators.canaug_traverse_edge(g, aut_gens, property, dig=False, loops=False, sparse=True)#
Main function for exhaustive generation. Recursive traversal of a canonically generated tree of isomorph free graphs satisfying a given property.
INPUT:
g- current position on the tree.aut_gens- list of generators of Aut(g), in list notation.property- check before traversing below g.
EXAMPLES:
sage: from sage.graphs.graph_generators import canaug_traverse_edge sage: G = Graph(3) sage: list(canaug_traverse_edge(G, [], lambda x: True)) [Graph on 3 vertices, ... Graph on 3 vertices]
The best way to access this function is through the graphs() iterator:
Print graphs on 3 or less vertices.
sage: for G in graphs(3): ....: print(G) Graph on 3 vertices Graph on 3 vertices Graph on 3 vertices Graph on 3 vertices
Print digraphs on 3 or less vertices.
sage: for G in digraphs(3): ....: print(G) Digraph on 3 vertices Digraph on 3 vertices ... Digraph on 3 vertices Digraph on 3 vertices
- sage.graphs.graph_generators.canaug_traverse_vert(g, aut_gens, max_verts, property, dig=False, loops=False, sparse=True)#
Main function for exhaustive generation. Recursive traversal of a canonically generated tree of isomorph free (di)graphs satisfying a given property.
INPUT:
g- current position on the tree.aut_gens- list of generators of Aut(g), in list notation.max_verts- when to retreat.property- check before traversing below g.degree_sequence- specify a degree sequence to try to obtain.
EXAMPLES:
sage: from sage.graphs.graph_generators import canaug_traverse_vert sage: list(canaug_traverse_vert(Graph(), [], 3, lambda x: True)) [Graph on 0 vertices, ... Graph on 3 vertices]
The best way to access this function is through the graphs() iterator:
Print graphs on 3 or less vertices.
sage: for G in graphs(3, augment='vertices'): ....: print(G) Graph on 0 vertices Graph on 1 vertex Graph on 2 vertices Graph on 3 vertices Graph on 3 vertices Graph on 3 vertices Graph on 2 vertices Graph on 3 vertices
Print digraphs on 2 or less vertices.
sage: for D in digraphs(2, augment='vertices'): ....: print(D) Digraph on 0 vertices Digraph on 1 vertex Digraph on 2 vertices Digraph on 2 vertices Digraph on 2 vertices
- sage.graphs.graph_generators.check_aut(aut_gens, cut_vert, n)#
Helper function for exhaustive generation.
At the start, check_aut is given a set of generators for the automorphism group, aut_gens. We already know we are looking for an element of the auto- morphism group that sends cut_vert to n, and check_aut generates these for the canaug_traverse function.
EXAMPLES:
Note that the last two entries indicate that none of the automorphism group has yet been searched - we are starting at the identity [0, 1, 2, 3] and so far that is all we have seen. We return automorphisms mapping 2 to 3:
sage: from sage.graphs.graph_generators import check_aut sage: list( check_aut( [ [0, 3, 2, 1], [1, 0, 3, 2], [2, 1, 0, 3] ], 2, 3)) [[1, 0, 3, 2], [1, 2, 3, 0]]
- sage.graphs.graph_generators.check_aut_edge(aut_gens, cut_edge, i, j, n, dig=False)#
Helper function for exhaustive generation.
At the start, check_aut_edge is given a set of generators for the automorphism group, aut_gens. We already know we are looking for an element of the auto- morphism group that sends cut_edge to {i, j}, and check_aut generates these for the canaug_traverse function.
EXAMPLES:
Note that the last two entries indicate that none of the automorphism group has yet been searched - we are starting at the identity [0, 1, 2, 3] and so far that is all we have seen. We return automorphisms mapping 2 to 3:
sage: from sage.graphs.graph_generators import check_aut sage: list( check_aut( [ [0, 3, 2, 1], [1, 0, 3, 2], [2, 1, 0, 3] ], 2, 3)) [[1, 0, 3, 2], [1, 2, 3, 0]]