MATLAB Answers

Adam
0

keep repeated x values and average y values

Adam
さんによって質問されました 2014 年 8 月 6 日
最新アクティビティ Nade Sritanyaratana さんによって コメントされました 2014 年 8 月 7 日
Dear all
I have the following data:
x=[4.1 2.7 3.2 2.5 4.1 2.5 0 5 6.1 2.5];
y=[10 16 11 24 8 18 20 7 6 4];
I would like to take repeated values only once that mean x should become:
xn=
0
2.5
2.7
3.2
4.1
5.
6.1
the correspodnig values of y should be the average:
yn=
20
15.33333333
16
11
9
7
6
that means 4.1 is rpeated twice: x(1)=4.1 x(5)=4.1
and thus the corresponding y data are: y(1)=10 y(5)=8
in this case only one 4.1 value is keept but y should be averaged: (10+8)/2 = 9
your helps are highly appreciated.
cheers

  0 件のコメント

サインイン to comment.

1 件の回答

回答者: Nade Sritanyaratana 2014 年 8 月 6 日

You may want to consider using accumarray .
The following code uses unique to find all unique values of x and also an inverse mapping ix from the unique array to the original array index. It then uses accumarray to average all y data that has matching ix:
x=[4.1 2.7 3.2 2.5 4.1 2.5 0 5 6.1 2.5];
y=[10 16 11 24 8 18 20 7 6 4];
[xn, ~, ix] = unique(x)
yn = accumarray(ix,y,[],@mean)

  4 件のコメント

Nade Sritanyaratana 2014 年 8 月 6 日
Hi,
I don't get this error just by running the exact lines I provided, but it sounds like for some reason ix and y do not have the same number of elements in your environment. Try looking at the size of both arrays:
>> size(y)
>> size(ix)
Adam
2014 年 8 月 7 日
Hi,
this eamaple is not working:
x=[4.1 2.7 3.2 2.5 4.1 2.5 0 5 6.1 2.5];
y=[10 16 11 24 8 18 20 7 6 4];
[xn, ~, ix] = unique(x)
cheers yn = accumarray(ix,y,[],@mean)
ix and y in this example have the same size i.e. 10
Nade Sritanyaratana 2014 年 8 月 7 日
I am not getting the same error using MATLAB R2014a. It's possible that one or more of the functions (unique, accumarray, or mean) are being shadowed. You can check this by running "which -all {function_name}" on the command window.

サインイン to comment.



Translated by