''Center'' value is larger than the number of nodes in the graph ({0}). Multigraph not supported. Source nodes must be unique. Target nodes must be unique. Input matrices must have equal number of columns. Source and target nodes must have the same number of elements, unless one is a scalar. Input must be a sparse matrix. Input must be a square sparse matrix with real double elements. Flag ''transp'' only supported for sparse matrices. Flag ''transp'' only supported for directed graph. Incidence matrix not supported for graphs with self-loops. G1 and G2 must have the same number of nodes and edges. Nodes in the same category must have the same degree, and the number of nodes in each category must be the same in G1 and G2. Invalid data type. Adjacency matrix must be a real square matrix which is double, single, or logical. When specified as a number, ''Center'' value must be a positive integer. ''Center'' value must be a numeric node index or a node name. Shortest path to nearest nodes in a subset unavailable EdgeCats must be a matrix of size numedges(G)x2 and of type uint64. EdgeCats must be a matrix of type uint64, of size either numedges(G)x2 or 0x0. NodeCats must be a matrix of size numnodes(G)x2 and of type uint64. NodeCats must be integers between 0 and N-1, with N the number of categories. Option must be ''transp''. Table format must be a logical scalar. Must be true when more than one event is set. ''UseGravity'' value must be ''on'' or ''off''. isInnerNode must be a dense logical vector of length n. Number of iterations must be a finite positive integer-valued scalar. Event selection vector must be a six element logical vector. Layer size must be a dense double vector of length at least 1 containing positive integers. Xcoord must be a real dense double vector of length equal to the number of nodes. LeftRight must be a logical scalar. Parameter value must be a nonnegative integer or Inf. Parameter value must be a positive integer or Inf. Reachability must be a dense logical vector with length equal to the number of nodes. Number of nodes must be a positive integer. Node ID must be a positive integer scalar not greater than the number of nodes in the graph ({0,number,integer}). Node IDs must be a double array of positive integers not greater than the number of nodes in the graph ({0,number,integer}). Number of nodes must be a real, nonnegative integer. Restart value must be true or false. Source must be a numeric array of positive integer node indices. Source node IDs must be positive integers not greater than the number of nodes in the graph ({0,number,integer}). Source must be a vector of numeric node indices, or the character vector ''all''. Start node must be a double scalar node index. Subset elements must be valid numeric node indices. Allowed syntaxes are adjacency(G), adjacency(G, w) and adjacency(..., ''transp''), where w must be a real double vector of length numedges(G). Target must be a numeric array of positive integer node indices. Target node IDs must be positive integers not greater than the number of nodes in the graph ({0,number,integer}). Target must be a vector of numeric node indices, or the character vector ''all''. Target node must be a double scalar node index. UpperLower must be a logical scalar. Graph edge weights must be a double or single, real vector with length equal to the number of edges in the graph. Weights must be a dense real vector of type double and of length numedges(G). When ''WeightEffect'' value is ''direct'', edge weights must be nonzero and non-NaN. When ''WeightEffect'' value is ''inverse'', edge weights must be finite. ''WeightEffect'' value must be ''direct'',''inverse'', or ''none''. Initial X coordinates must be a real numeric vector with {0,number,integer} elements. Initial Y coordinates must be a real numeric vector with {0,number,integer} elements. Initial Z coordinates must be a real numeric vector with {0,number,integer} elements. The source and target nodes must be different. The maximum flow problem is badly scaled. Consider rescaling the edge weights. Input must be an MLDigraph. Input must be an MLGraph. Graph edge weights contain NaN. Graph edge weights must be nonnegative. Initial X coordinates must be finite. Initial Y coordinates must be finite. Initial Z coordinates must be finite. The graph adjacency matrix must be symmetric. Input graph has a two-cycle with two non-zero weights. Input graph must not have any self-loops. Digraph must be in reverse topological order. Number of nodes must be greater than or equal to {0,number,integer}. Adjacency matrix must be symmetric. ''UseGravity'' value must be ''on'', ''off'', or a logical scalar. Layout algorithm failed due to numeric overflow. Consider rescaling the graph edge weights. Sum of layer sizes must be equal to number of nodes. With one-input syntax, input must be dense and square. With two-input syntax, the first input must be sparse and square and the second input must be a real double vector of length equal the number of nonzeros of the first input. Unable to load graph or digraph object because it was corrupted.