Table - conditional maths and plots from variables

Dear all,

I have a 57254x87 table, on which I need to perform some maths and plots with the tables. I have one column called Subject_Id (total of 8 particpants) and another one called Trial_Id (a total of 6 per particpant), and I need to plot the data from two others variables. For example, I am looking for a code that will take the first subject, then run a loop for each one of the 6 trials, ploting the data of my columns 3 and 4, and then take the second subject, and reproduces the same thing one more time, and so on for the 8 subjects.

I guess I need one loop inside the other, but I am not sure how to do it.

Any help will be very much appreciated.

[Chunru]

post a portion of your data.

[Thiago de Aquino Costa Sousa]

@Chunru please considerer the number of the columns, because I was not able to copy the name of th variables. Subject_Id (column 86) Trial_Id (column 87) Xdisplacement (column 3) Ydisplacement (column 4). Thank you for your help.

             
914371	0.00356000000000000	-0.448190000000000	-3.46237000000000	161.552200000000	-0.0869400000000000	0.360670000000000	88.6375000000000	109.142620000000	514.040590000000	95.0959900000000	0	539.880920000000	83.5962300000000	-103.920560000000	504.995670000000	86.7854700000000	1	1	-3.46237000000000	0.670540000000000	540.376160000000	139.983540000000	0	539.880920000000	83.5962300000000	15.9204300000000	540.506470000000	133.368580000000	21.9610100000000	539.172970000000	108.898650000000	24.9944700000000	529.957210000000	85.0114200000000	10.8748500000000	535.254210000000	2.82422000000000	-17.1000100000000	543.319520000000	132.028410000000	-20.8233700000000	541.126340000000	106.628250000000	-23.7434300000000	528.965700000000	83.6034500000000	-6.90171000000000	534.684630000000	3.54805000000000	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	160.581540000000	0	0	-1.00000000000000e-05	0	0	160.581540000000	0	0	-1.00000000000000e-05	'Empty'	'None'	'HTY01'	'Trial_1_2Mi'
914372	0.00912000000000000	-0.448190000000000	-3.46237000000000	161.552200000000	-0.0869400000000000	0.360670000000000	88.6375000000000	109.142620000000	514.040590000000	95.0959900000000	0	539.880920000000	83.5962300000000	-103.920560000000	504.995670000000	86.7854700000000	1	1	-3.46237000000000	0.670540000000000	540.376160000000	139.983540000000	0	539.880920000000	83.5962300000000	15.9204300000000	540.506470000000	133.368580000000	21.9610100000000	539.172970000000	108.898650000000	24.9944700000000	529.957210000000	85.0114200000000	10.8748500000000	535.254210000000	2.82422000000000	-17.1000100000000	543.319520000000	132.028410000000	-20.8233700000000	541.126340000000	106.628250000000	-23.7434300000000	528.965700000000	83.6034500000000	-6.90171000000000	534.684630000000	3.54805000000000	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	160.581540000000	0	0	-1.00000000000000e-05	0	0	160.581540000000	0	0	-1.00000000000000e-05	'Empty'	'None'	'HTY01'	'Trial_1_2Mi'
914373	0.0146700000000000	-0.424390000000000	-3.43538000000000	161.535980000000	-0.0347900000000000	0.281270000000000	88.6616000000000	109.142620000000	514.040590000000	95.0959900000000	0	539.880920000000	83.5962300000000	-103.920560000000	504.995670000000	86.7854700000000	2	1	-3.43538000000000	0.670540000000000	540.376160000000	139.983540000000	0	539.880920000000	83.5962300000000	15.9204300000000	540.506470000000	133.368580000000	21.9610100000000	539.172970000000	108.898650000000	24.9944700000000	529.957210000000	85.0114200000000	10.8748500000000	535.254210000000	2.82422000000000	-17.1000100000000	543.319520000000	132.028410000000	-20.8233700000000	541.126340000000	106.628250000000	-23.7434300000000	528.965700000000	83.6034500000000	-6.90171000000000	534.684630000000	3.54805000000000	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	160.581540000000	0	0	-1.00000000000000e-05	0	0	160.581540000000	0	0	-1.00000000000000e-05	'Empty'	'None'	'HTY01'	'Trial_1_2Mi'
914374	0.0202300000000000	-0.424390000000000	-3.43538000000000	161.535980000000	-0.0347900000000000	0.281270000000000	88.6616000000000	109.142620000000	514.040590000000	95.0959900000000	0	539.880920000000	83.5962300000000	-103.920560000000	504.995670000000	86.7854700000000	2	1	-3.43538000000000	0.670540000000000	540.376160000000	139.983540000000	0	539.880920000000	83.5962300000000	15.9204300000000	540.506470000000	133.368580000000	21.9610100000000	539.172970000000	108.898650000000	24.9944700000000	529.957210000000	85.0114200000000	10.8748500000000	535.254210000000	2.82422000000000	-17.1000100000000	543.319520000000	132.028410000000	-20.8233700000000	541.126340000000	106.628250000000	-23.7434300000000	528.965700000000	83.6034500000000	-6.90171000000000	534.684630000000	3.54805000000000	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	160.581540000000	0	0	-1.00000000000000e-05	0	0	160.581540000000	0	0	-1.00000000000000e-05	'Empty'	'None'	'HTY01'	'Trial_1_2Mi'

[Chunru]

Try to upload a mat file:

data = yourdata(1:200, :);      % 1:200 or other number to demonstrate your case

Answer 1

[Paul]

Check out splitapply

[Thiago de Aquino Costa Sousa]

Hey Paul, I went through this documentation, and I see that it can solve my problem. However, when I run this code my only output is one plot, even though the documentation says that for each group, it executes the function once. Can you help me on that? I would need to plot 48 times, once for each group, and I also need that for each subject, it holds the plot for add the 6 trials for each one of the 8 participants together. Here it is my code:

[G,subjects] = findgroups(masterTable.Subject_Id,masterTable.Trial_Id);% G has the groupos from a combination of masterTable.Subject_Id,masterTable.Trial_Id, subjects the name of the subjects many times as the number of trials.
    func = @(x,y) plot(x,y);
    figure
    xlim([-150 150]);ylim([-10 510])
    subj_trajectory = splitapply(func,masterTable.CameraX,masterTable.CameraY,G);% just return one plot, which one I cannot identify the the group originary.

[Adam Danz]

While splitapply certainly could be used, I don't think it's the best option. A simple for-loop (or two nested for-loops) would likely be faster and more readable.

[Thiago de Aquino Costa Sousa]

Hi @Adam Danz, thank you very much for you reply. I am pretty new in coding and Matlab. I have this code so far...

clear
clc
mytopfolder = '/Users/.../Participants data';
filepattern = fullfile(mytopfolder, '**/VIVE/Walking_ST/Trial_.txt');  
thefiles = dir(filepattern);
nfiles = length(thefiles);
for k = 1:nfiles
    basefilename = thefiles(k).name;
    fullfilename = fullfile(thefiles(k).folder, basefilename);
    fprintf(1, 'Now reading %s\n', fullfilename);
    thisTable = readtable(fullfilename);
    thisTable.Subject_Id(:) = {fullfilename(98:102)};
    thisTable.Trial_Id(:) = {fullfilename(120:130)};
    fprintf('    It has %d rows in it.\n', height(thisTable));
    if k == 1
        masterTable = thisTable;
    else
        masterTable = [masterTable; thisTable];
    end
    fprintf('    Now there are a total of %d rows in the master table.\n', height(masterTable));
end
masterTable.CameraX = masterTable.CameraX*-1;

And this is the structure of directory I have

Based on that, would you be able to hep me with this two nested for-loops as I don't have idea how to write it, that allows me to make a plot for each of my Subject associated to the trial Id. Thank you.

[Paul]

Two nested for loops, with a unique() inside each could definitely do the job. I guess I've gotten into the splitapply habit. Don't know which is faster, or which would be considered more readable.

The splitapply approach would go as follows (completely untested code)

splitapply(@(trialid,col3,col4) plottrajid(trialid,col3,col4),T.Trial_ID,T.col3,T.col4,findgroups(T.Subject_Id))

function plottrialid(trialid,col3,col4)
splitapply(@(col3,col4) plotcols(col3,col4),col3,col4,findgrous(trialid))
end

function plotcols(col3,col4)
figure
plot(col3,col4)
xlabel('whatever')
% etc
end

[Thiago de Aquino Costa Sousa]

Hey Paul. Thank you for replying to me. I corrected what I guess could be some typos, then the code comes to:

splitapply(@(trialid,col3,col4) plottrajid(trialid,col3,col4),T.Trial_ID,T.col3,T.col4,findgroups(T.Subject_Id))
function plottrialid(trialid,col3,col4)
splitapply(@(col3,col4) plotcols(col3,col4),col3,col4,findgrous(trialid))
end
function plotcols(col3,col4)
figure
plot(col3,col4)
xlabel('whatever')
% etc
end

However, when I run the code I got the following error:

Error using splitapply

Applying the function '@(trialid,col3,col4)plottrialid(trialid,col3,col4)' to the 1st group of data generated the

following error: Undefined function 'plottrialid' for input arguments of type 'cell'.

[Paul]

Only one splitapply is needed

Create some data in a table.

rng(100)
T = table;
T.subj = [ones(10,1); 2*ones(10,1)];
T.trial = 3*randi(2,20,1);
T.col3 = (1:20).';
T.col4 = T.subj + T.trial;

Make the plots

G = findgroups(T.subj,T.trial);
splitapply(@(subj,trial,col3,col4) makeplot(subj(1),trial(1),col3,col4),T.subj,T.trial,T.col3,T.col4,G);

function makeplot(subj,trial,col3,col4)
figure
plot(col3,col4,'-o')
title("subj = " + double(subj) + ", trial = " + double(trial));
end

[Thiago de Aquino Costa Sousa]

Hey @Paul, thank you for replying to me. I will plot my data on your code and check what happens next... sincerely, Thiago