Replace stale / repeating data with NaN in a table

I have data in a time table with around 10 variables. occasionally the data source "stalls out" and just keeps outputting the last value instead of fresh data in a particular column. The timestamp and some columns will update correctly. What I want to do is replace the stale / repeated values with NaN. I can't just remove the row as I need the other columns. When the data is good there would never be a case where the same value repeats twice or more in a row.

example input

data.Var1 = [ 3 , 2, 5, 4, 4 , 4, 4, 6, 2, 3 , 5, 5 ,5 ,5 ]

desired output

data.Var1 = [ 3 , 2, 5, 4, NaN , NaN, NaN, 6, 2, 3 , 5, NaN , NaN ,NaN ]

I saw something in a previous post (https://www.mathworks.com/matlabcentral/answers/216921-need-to-remove-repeated-adjacent-elements-in-an-array?s_tid=srchtitle ) that could work but I'm struggling to modify it for my needs as it removes values, once I put in NaN the test doesnt know where the last good value was.

I'd consider upgrading if theres a function in a newer version or toolbox.

Answer 1

[KSSV]

Var1 = [ 3 , 2, 5, 4, 4 , 4, 4, 6, 2, 3 , 5, 5 ,5 ,5 ] ;
% Var1 = [ 3 , 2, 5, 4, NaN , NaN, NaN, 6, 2, 3 , 5, NaN , NaN ,NaN ] 

idx = find(diff(Var1)==0)+1 ; 
Var1(idx) = NaN 

Var1 = 1×14

3 2 5 4 NaN NaN NaN 6 2 3 5 NaN NaN NaN

[Markus Niemelä]

Hi!

I was wondering, how to do this exact thing, but for multiple columns?

For example: Let's say I have matrix:

1 2 3 4

2 3 4 4

3 1 3 1

3 1 3 2

And the desired output would then be:

1 2 3 4

2 3 4 NaN

3 1 3 1

NaN NaN NaN 2

I hope this makes sense,

Kr, Markus

Answer 2

[Mohammad Sami]

You can try this.

Var1 = [ 3 , 2, 5, 4, 4 , 4, 4, 6, 2, 3 , 5, 5 ,5 ,5 ];
i = Var1(2:end) == Var1(1:end-1);
i = [false i];
Var1(i) = NaN

Var1 = 1×14

3 2 5 4 NaN NaN NaN 6 2 3 5 NaN NaN NaN