Why do I get this Error using . "Array indices must be positive integers or logical values."

This is my code:

%% 
 %extract files that have '21' in them
files = dir("*21*");

S = struct;

figure(1);clf
t1 = tiledlayout('flow');

for k = 1:length(files)
  
    T_name = ['Exp_' + string(files(k).name)];
    S.(T_name) = readtable(files(k).name);

    nexttile
    plot(S.(T_name).Time_s_, S.(T_name).AI0HPF_V_)
    title(strrep(T_name, '_', ' '));
    axis([-inf, inf, -1.5, 1.5]);
   
    clear T_name
end

t1.Padding = 'compact';
t1.TileSpacing = 'compact';


%% 
T_qemg = table;
T_qemg.lb_0 = NaN(5,1);
T_qemg.lb_5 = NaN(5,1);
T_qemg.lb_10 = NaN(5,1);
T_qemg.lb_15 = NaN(5,1);
T_qemg.lb_20 = NaN(5,1);

names2 = fieldnames(S);

for b = 1:numel(names2)
    REMG = S.(string(names2(b))).AI0HPF_V_;
    rec_EMG = abs(REMG(3000:5000));
    RMS = sqrt(movmean(rec_EMG.^2,200));

    %S.(string(names2(b))).RMS = RMS;

    QEMG = mean(RMS);
    %variance = var(REMG(2000:4000));
    %QEMG = mean(variance);

    parts = strsplit(string(names2(b)),'_');
    weight = str2double(parts(3));
    row = str2double(parts(4));
   
    if parts(3) == "0"
        T_qemg.lb_0(row)=QEMG;
    elseif parts(3) == "5"
        T_qemg.lb_5(row)=QEMG;
    elseif parts(3) == "10"
        T_qemg.lb_10(row)=QEMG;
    elseif parts(3) == "15"
        T_qemg.lb_15(row)=QEMG;
    elseif parts(3) == "20"
        T_qemg.lb_20(row)=QEMG;
    else
        fprintf("data error in file names")
    end
    clear REMG variance QEMG parts row weight
end

figure(2);clf

p= plot(table2array(T_qemg)','o');

And this is the error message I can't seem to get past:

Error using .

Array indices must be positive integers or logical values.

Error in tutorialplots (line 72)

T_qemg.lb_0(row)=QEMG;

The first section of code runs fine, it's only second section I'm struggling with. "T_name" is of the form "Exp_21_5_e" so "parts" is a 1x4 string.

Answer 1

[Walter Roberson]

The code splits apart the name at '_" characters, so for the name "Exp_21_5_e" parts(1) would be "Exp", parts(2) would be "21", parts(3) would be "5" and parts(4) would be 'e'

It then does a str2double(parts(4)) would attempts to convert the string 'e' to a number. The result of that is going to be NaN.

It then tries to use that NaN as a row index.

You should be check isnan(Row) and if so do something appropriate.