'Grid Vectors not strictly monotonic increasing'

Hi,

I am trying to interpolate 3 datasets using interp1...but am getting the error 'Grid Vectors not strictly monotonic increasing' - which I do not understand.

My datasets are given in the attachement. I want to interpolate each to find at what value of A each of C11, C12 and C44 equal a known reference values - C11ref, C12ref, C44ref. I have used the following.

C11A = interp1(C11, A, C11ref);
C12A = interp1(C12, A, C12ref);
C44A = interp1(C44, A, C44ref);

The first of these works fine. The error is thrown for the second and third...

If anybody could shed some light, I would be grateful.

Answer 1

[Jos (10584)]

This happens when values in X are not unique

X = [1 2 3 3 4 5]
Y = [10 20 28 32 40 50]
interp1(X,Y, 3.5)

A workaround is to accumulate Y over all unique values of X

X = [1 2 3 3 4 5]
Y = [10 20 28 32 40 50]
[X2, ~, jx] = unique(X)
Y2 = accumarray(jx, Y ,[], @mean)
R = interp1(X2,Y2, 3.5)

[Guillaume]

This happens when values in X are not unique. No, this happens when the values is X are not strictly monotonic increasing as per the error message. If

X = [1 3 5 4 2]

between which indices of X should the interpolation of 2.5 occur?

[Jos (10584)]

My sentence was incorrect. I have edited it.

interp1 sorts the X values, and then the Y values accordingly

X = [1 2 3 4 5] ; Y = [10 20 30 40 50] ;
ri = randperm(5)
interp1(X(ri),Y(ri),2.5) %→ 25, always

Answer 2

[umesh sahu]

what if i want to delete all the values for X=3 and carry out the interpolation with rest X values ,how the "accumarray" command will have to change?