Omit marker NaN data

Sorry I am very new to MATLAB. I have 2 truct arrays where I need to truncate NaN values from the arrays. About the first 260 rows are NaN values. How would I go about doing this?

a =

struct with fields:

LANK: [818×3 double]

LGTRO: [818×3 double]

LHEE: [818×3 double]

LKNE: [818×3 double]

LSHO: [818×3 double]

LTOE: [818×3 double]

RANK: [818×3 double]

RGTRO: [818×3 double]

RHEE: [818×3 double]

RKNE: [818×3 double]

RSHO: [818×3 double]

RTOE: [818×3 double]

STRNM: [818×3 double]

nFrame: [818×1 double]

time: [818×1 double]

b =

struct with fields:

LANK: [797×3 double]

LGTRO: [797×3 double]

LHEE: [797×3 double]

LKNE: [797×3 double]

LSHO: [797×3 double]

LTOE: [797×3 double]

RANK: [797×3 double]

RGTRO: [797×3 double]

RHEE: [797×3 double]

RKNE: [797×3 double]

RSHO: [797×3 double]

RTOE: [797×3 double]

STRNM: [797×3 double]

nFrame: [797×1 double]

time: [797×1 double]

[KSSV]

Read about isnan.

[Walter Roberson]

rmmissing()

Answer 1

[Voss]

a = struct('LANK',[NaN(260,3); rand(558,3)],'LGTRO',[NaN(260,3); rand(558,3)])

a = struct with fields:
     LANK: [818×3 double]
    LGTRO: [818×3 double]

f = fieldnames(a);
% go through each field of struct 'a'
for ii = 1:numel(f)
    % find the index of the first row of 
    % field f{ii} in 'a' that is not all NaN:
    idx = find(~all(isnan(a.(f{ii})),2),1);
    if isempty(idx)
        % no such row means field f{ii} is all NaNs -> remove the whole
        % thing by treating it as if it has a row with a non-NaN value one
        % row beyond its last row:
        idx = size(a.(f{ii}),1)+1;
    end
    % remove rows up to and including row idx-1:
    a.(f{ii})(1:idx-1,:) = [];
end
a

a = struct with fields:
     LANK: [558×3 double]
    LGTRO: [558×3 double]