why do I get "Array indices must be positive integers or logical values"

D = diabetesdata;

Dm = mean (D);

Dm = 3.8451 120.8945 69.1055 20.5365 79.7995 31.9926 0.4719 33.2409 0.3490

Dsm = D - Dm

Dsm =

0.1549 20.1055 4.8945 -20.5365 -79.7995 -4.3926 -0.2279 6.7591 -0.3490

3.1549 73.1055 -1.1055 7.4635 -79.7995 3.9074 0.2731 7.7591 0.6510

4.1549 60.1055 -1.1055 15.4635 415.2005 -1.8926 0.1431 26.7591 0.6510

-2.8451 7.1055 28.8945 20.4635 -21.7995 0.0074 0.8491 -0.2409 0.6510

4.1549 -11.8945 6.8945 18.4635 34.2005 -4.0926 0.1681 -2.2409 0.6510

1.1549 18.1055 10.8945 14.4635 80.2005 -0.3926 -0.1109 -8.2409 0.6510

-0.8451 -9.8945 -7.1055 -20.5365 -79.7995 -9.3926 -0.3299 -12.2409 -0.3490

5.1549 31.1055 8.8945 13.4635 91.2005 2.2074 0.4211 -0.2409 0.6510

3.1549 57.1055 14.8945 -20.5365 -79.7995 7.9074 -0.1409 7.7591 0.6510

-2.8451 9.1055 0.8945 -7.5365 25.2005 -6.0926 0.0001 -11.2409 -0.3490

-2.8451 -25.8945 4.8945 0.4635 -6.7995 -6.0926 0.2011 2.7591 -0.3490

.....

Dstd = std(D)

Dstd = 3.3696 31.9726 19.3558 15.9522 115.2440 7.8842 0.3313 11.7602 0.4770

Dnorm = Dsm./Dstd(D)

Array indices must be positive integers or logical values.

why am i getting this when its always worked before?

Answer 1

[Torsten]

Most probably you mean

Dnorm = Dsm./Dstd

instead of

Dnorm = Dsm./Dstd(D)

[huan]

THE SIZE OF D IS LARGER THAN DSTD, AND Dstd(D) IS INCORRECT

[Torsten]

THE SIZE OF D IS LARGER THAN DSTD

It's enough that the number of columns of D and Dstd are the same.

A = rand(3,6)

A = 3×6

0.9177 0.9634 0.3477 0.4052 0.2695 0.2948 0.5912 0.1184 0.9285 0.9043 0.7753 0.5850 0.3678 0.9762 0.9808 0.1172 0.2459 0.3344

b = rand(1,6)

b = 1×6

0.6558 0.8259 0.9733 0.3385 0.3217 0.8282

Abyb = A./b

Abyb = 3×6

1.3994 1.1666 0.3572 1.1973 0.8377 0.3559 0.9015 0.1434 0.9540 2.6718 2.4103 0.7064 0.5609 1.1820 1.0077 0.3462 0.7644 0.4038

[Image Analyst]

Perhaps you need:

Dstd = std(D(:))

or

Dstd = std(D, "all")

[Torsten]

I don't think so. Mean and Standard deviation are taken columnwise for the matrix D. Finally, D is normalized to Dnorm. The only problem is the use of

Dnorm = Dsm./Dstd(D)

instead of

Dnorm = Dsm./Dstd

[Georgia Thorpe]

thank you it was that which was the propblem

Answer 2

[Walter Roberson]

Dstd(D) is requesting to index the vector Dstd by the array D

Remember MATLAB has absolutely no implied multiplication.

The only implied operations that I can think of at the moment are:

• the name of a function that is not in quotes or comments, and is not proceeded by @ or ? and is not followed by (, is treated as an implicit call to the function as if () had been coded after the name
• inside [] and {} then space is horizontal concatenation
• inside [] and {} then end of line is vertical concatenation
• in the symbolic toolbox, inside piecewise() calls, A RELATION B RELATION C is sometimes treated as A RELATION B & B RELATION C

Answer 3

[Image Analyst]

To learn about this frequently asked question, see the FAQ:

https://matlab.fandom.com/wiki/FAQ#%22Subscript_indices_must_either_be_real_positive_integers_or_logicals.%22