All -9999 in a table to NaN

2 views (last 30 days)
Huw Wadkin
Huw Wadkin on 27 Aug 2021
Commented: Huw Wadkin on 27 Aug 2021
Hi Everyone,
I have a table with results from a lot of sensors (1 column per sensor), if the results aren't valid then the output value is -5555.
I want to make all values of -5555 in the entire table =NaN, is there a quick way to do this?
I can do it for 1 column and create a loop to do it for all columns but is there an easier way?
Thanks in advance!

Accepted Answer

Steven Lord
Steven Lord on 27 Aug 2021
Use standardizeMissing.
dblVar = [NaN;3;Inf;7;9];
cellstrVar = {'one';'three';'';'N/A';'nine'};
charVar = ['A';'C';'E';' ';'I'];
categoryVar = categorical({'red';'yellow';'blue';'violet';''});
A = table(dblVar,cellstrVar,charVar,categoryVar)
A = 5×4 table
dblVar cellstrVar charVar categoryVar ______ __________ _______ ___________ NaN {'one' } A red 3 {'three' } C yellow Inf {0×0 char} E blue 7 {'N/A' } violet 9 {'nine' } I <undefined>
B = standardizeMissing(A, 7)
B = 5×4 table
dblVar cellstrVar charVar categoryVar ______ __________ _______ ___________ NaN {'one' } A red 3 {'three' } C yellow Inf {0×0 char} E blue NaN {'N/A' } violet 9 {'nine' } I <undefined>
  1 Comment
Huw Wadkin
Huw Wadkin on 27 Aug 2021
Thanks You! I knew I was missing something simple!

Sign in to comment.

More Answers (1)

dpb
dpb on 27 Aug 2021
isBad=(T{:,:)==-5555); % logical array across table T
T{:,:)(isBad)=nan; % assign NaN to those locations
Use your table variable name in place of T above, of course.
Depending on how you're creating the table, it may be easier/more direct to use the IsMissing indicator when create the table rather than have to fix up after the fact; we would have to know how the table is being populated to begin with to say -- that is, we need to see the code.

Categories

Find more on Numeric Types in Help Center and File Exchange

Tags

Products


Release

R2019b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!