How to average finite components only of 4 data?

Question

Yandong Lang 2019-3-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/451091-how-to-average-finite-components-only-of-4-data

回答： Stephan 2019-3-20

My question is that I have four 3D ocean data(360*160*50). I want to average the finite numbers of the 4 data. For example:

data1(1,2,3) = NaN;

data2(1,2,3) = 3;

data3(1,2,3) = NaN;

data4(1,2,3) =4

In this case, I will sum data2(1,2,3) and data4(1,2,3) and then dividid by 2. How to do it for the whole data?

It needs to be taken into account that continent area of the 4 data are all NaN because it is ocean data. Due to this, I can not replace all the NaN to be zero.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Stephan 2019-3-20

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/451091-how-to-average-finite-components-only-of-4-data#answer_366358

Hi,

the mean function has the option nanflag. That makes life easy:

% create example data containing NaN values
data1 = randi(100,5,4,3);
data2 = randi(100,5,4,3);
data3 = randi(100,5,4,3);
data1(randperm(60,15))= NaN;
data2(randperm(60,15))= NaN;
data3(randperm(60,15))= NaN;
% calculate the average ignoring NaN-data 
result = mean(reshape(vertcat(data1,data2,data3),[],1),'omitnan')