histcount does not work as excepted

Dear all,

I want to calculate the position of double (or multiple) values inside a list. I found an answer in this forum, however I have stumbled across a problem with histcount. Is the following behaviour of histcount as excepted. The function histc works perfectly.

a=[1 2 3 4];
b=[1 1 2 2 3 3 4 4];
[c,d]=histcounts(b,a);
multi=sum(b==d(c>=2)')>=1
% 1   1   1   1   1   1   0   0
[c,~]=histc(b,a);
multi=sum(b==a(c>=2)')>=1
% 1   1   1   1   1   1   1   1

The algorithm is based on:

https://www.mathworks.com/matlabcentral/answers/336500-finding-the-indices-of-duplicate-values-in-one-array

I use Matlab 2018a on a Ubuntu 20.04 LTS system.

Answer 1

[Steven Lord]

From the documentation page:

The value X(i) is in the kth bin if edges(k) ≤ X(i) < edges(k+1). The last bin also includes the right bin edge, so that it contains X(i) if edges(end-1) ≤ X(i) ≤ edges(end).

The first bin counts values in b that satisfy 1 <= b < 1

The last bin counts values in b that satisfy 3 <= b <= 4.

Thus histcounts is behaving as expected.

histc had an extra bin that counted exactly those values that matched the last element of the edges.

a=[1 2 3 4];
b=[1 1 2 2 3 3 4 4];
[c,d]=histcounts(b,a) % 3 bins

c = 1×3

2 2 4

d = 1×4

1 2 3 4

[c2,d2]=histc(b,a) % 4 bins

c2 = 1×4

2 2 2 2

d2 = 1×8

1 1 2 2 3 3 4 4

[c3, d3] = histcounts(b, [a Inf]) % 4 bins

c3 = 1×4

2 2 2 2

d3 = 1×5

1 2 3 4 Inf

In that last histcounts call the next-to-last bin counts elements of b that satisfy 3 <= b < 4 and the last bin those that satisfy 4 <= b <= Inf. I could have used any value greater than or equal to 4 as that last bin edge, but Inf guarantees you catch any non-NaN real values in your data no matter how large they are.

If you're looking to find where certain values are located in a list, perhaps the ismember function would be more suitable for your application?

[ConvexHull]

Thank you!