max

Hi,

I'm selecting the max of three different values

 distmax=max([dist1 dist2 dist3])

which gives for example

 distmax= 3.434

is there a possibility to have the name of the value instead of the value itself, like

 distmax = dist1

thank you

Answer 1

[Walter Roberson]

No.

You can, however, use

[distmax, idx] = max([dist1 dist2 dist3])

and then idx will indicate which index in to the [dist1 dist2 dist3] matrix was the maximum. To figure out which of the variables that came from, you would need to know the size()'s of the variables.

It would be easier if you used

[distmax, idx] = max([max(dist1(:)), max(dist2(:)), max(dist3(:))])

and then idx would indicate which variable number the max was found in; relating the variable number to the variable name is a different matter.

Plain

[distmax, idx] = max([dist1 dist2 dist3])

is the abbreviation for the above for the special case where dist1, dist2, and dist3 are all guaranteed to be scalars (something your question does not allow us to assume).

Question: what do you want to happen if the identical maximum value exists in more than one of the variables?

[Nicolas]

thanks for your answer, actually my problem is a litle bit different.

I'm first calculating distances between 3 different points:

distij=sqrt((x(j)-x(i))^2+(y(j)-y(i))^2);

distik=sqrt((x(k)-x(i))^2+(y(k)-y(i))^2);

distjk=sqrt((x(k)-x(j))^2+(y(k)-y(j))^2);

then i want to get the maximum of those distances.

max([distij distik distjk]);

I was asking how to extract the name because I would like to calculate the barycenter of my line. so I need to know the two extrem points of the line either x(i,j or k) and y(i,j or k).

does that make sense?

[Walter Roberson]

v = [j i;k i;k j];

[distmax, idx] = max([distij distik distjk]);

xextr = x(v(idx,:));

yextr = y(v(idx,:));

First extreme would be xextr(1), yextr(1), second would be xextr(2), yextr(2)

[Nicolas]

great ! thanks a lot

Answer 2

[Jan]

It is possible to get the name:

a = rand;
b = rand;
c = rand;

function Name = GetMaxName(a, b, c)
[Value, Index] = max([a, b, c]);
Name = inputname(Index);

But Wlater's solution is much more stable and reliable, because INPUTNAME replies an empty string for temporary objects, e.g. in "GetMaxName(round(a) ...)". Using the name of a variable as information is a fragil programming method.