Error when running G= graph(s,t) in matlab

I want to calculate L = laplacian(G) from a graph dataset. I imported the dataset, from here , which contains two columns as shown below:

# Nodes: 3997962 Edges: 34681189
# FromNodeId    ToNodeId
0   1
0   2
0   31
0   73
0   80
0   113619
0   2468556
0   2823829
0   2823833
0   2846857
0   2947898
0   3011654
0   3701688
0   3849377
0   4036524
0   4036525
0   4036527
0   4036529
0   4036531
0   4036533
0   4036534
0   4036536
0   4036537
1   2
1   3
1   4
1   5
1   6
1   7
1   8
1   9
1   10
1   11

To do so, I need to find G first so I use G = graph(FromNodeId, ToNodeId). When I did that, I got this error:

>> G = graph(fromNodeId,toNodeId)
Error using matlab.internal.graph.MLGraph
Source must be a dense double array of node indices.

Error in matlab.internal.graph.constructFromEdgeList (line 125)
G = underlyingCtor(double(s), double(t), totalNodes);

Error in graph (line 264)
                matlab.internal.graph.constructFromEdgeList(...

I don't know why! Can I get a solution of that? Thank you.

[Walter Roberson]

how did you import the Data and construct the node id vectors?

[Ahmad Aseeri]

I used (import data) from Matlab itself and chose (column vectors)

Answer 1

[Steven Lord]

Indices in MATLAB start at 1, not 0. Try adding 1 to your index vectors.

[Ahmad Aseeri]

Wow! You are right. Thank you vry much

Answer 2

[lol]

THIS IS MY CODE:

filename = 'TABLA_AF1';
s = xlsread(filename,'A2:A25');
t = xlsread(filename,'B2:B25');
weights = xlsread(filename,'C2:C25');
G = graph(s,t,weights);
p = plot(G,'EdgeLabel',G.Edges.Weight);
[T,pred] = minspantree(G);
highlight(p,T)

AND THIS IS MY BASE DATA:

Nodo Inicial Nodo Final Distancia

A B 2

A E 16

A C 3

C F 3

C I 12

F E 7

F I 9

F H 6

I H 14

E H 2

E G 4

B E 8

B D 11

D E 2

D G 5

D J 7

G J 1

G L 6

H G 3

J L 12

H L 17

I K 7

H K 9

K L 17

AND I GET THIS ERROR:

Error using matlab.internal.graph.constructFromEdgeList (line 249)

Weight must have as many elements as edge list or be a scalar.

Can you tell me why please?

[Walter Roberson]

What is reported for size(s), size(t), size(weights)? Also what is class() of each of those?

[lol]

how can I check those parameters?

[Walter Roberson]

at the command prompt

size(s), size(t), size(weights)
class(s), class(t), class(weights)

[lol]

I just get these:

weightsError using matlab.internal.graph.constructFromEdgeList (line 249)

Weight must have as many elements as edge list or be a scalar.

Error in graph (line 300)

matlab.internal.graph.constructFromEdgeList(...

Error in A01382540_AF1 (line 5)

G = graph(s,t,weights);

[Walter Roberson]

These are not commands intended to somehow directly make the code work. These are commands to execute at the command line after the error, and then you post what the results were for us to examine so that we can make better guesses about what the problem is.

[Steven Lord]

You asked this as a separate question: https://www.mathworks.com/matlabcentral/answers/1583024-error-using-matlab-internal-graph-constructfromedgelist-line-249-weight-must-have-as-many-elements?s_tid=srchtitle

[lol]

filename = 'TABLA_AF1.xlsx';
s = xlsread(filename,'A2:A25');
t = xlsread(filename,'B2:B25');
weights = xlsread(filename,'C2:C25');
G = graph(s,t,weights);
p = plot(G,'EdgeLabel',G.Edges.Weight);
[T,pred] = minspantree(G);
highlight(p,T)

[lol]

YES PLEASE, I NEED HELP

[Walter Roberson]

Execute this code:

filename = 'TABLA_AF1.xlsx';
s = xlsread(filename,'A2:A25');
t = xlsread(filename,'B2:B25');
weights = xlsread(filename,'C2:C25');
size(s), size(t), size(weights)
class(s), class(t), class(weights)

and tell us what the output is.

[lol]

>> A01382540_AF1

ans =

0 0

ans =

0 0

ans =

24 1

ans =

'double'

ans =

'double'

ans =

'double'

[Walter Roberson]

filename = 'TABLA_AF1.xlsx';
[~, s] = xlsread(filename,'A2:A25');
[~, t] = xlsread(filename,'B2:B25');
weights = xlsread(filename,'C2:C25');
G = graph(s,t,weights);
p = plot(G,'EdgeLabel',G.Edges.Weight);
[T,pred] = minspantree(G);
highlight(p,T)

When you use xlsread() then the first output is always purely numeric. But you are trying to read a text column for s and t, and the numeric interpretation of those is all NaN. It happens that xlsread() trims out leading and trailing NaN columns and rows, so the column of NaN gets trimmed away entirely leaving you with empty entries for s and t.

My adjusted code here uses the second output instead, which is the text entries.

You would be better of using readtable(), which is able to automatically detect the data type of inputs.

[lol]

Thank you so much! it already works!