Undefined function 'C' for input arguments of type 'double'.

hello!

i am running into a problem with my script with this line:

whole_ts = {1:2150, 1:1928, 1:2426};

I've tried:

whole_ts = [{1:2150, 1:1928, 1:2426}];
whole_ts = (1:2150, 1:1928, 1:2426);
whole_ts = [C{1:2150, 1:1928, 1:2426}];
whole_ts = timeseries([1:2150], [1:1928], [1:2426])

this one seems the most promising:

whole_ts = [C{1:2150, 1:1928, 1:2426}];

where i am getting this error:

Undefined function 'C' for input arguments of type 'double'.

I am trying to load it into a cell like I did with the 'trs' variable which is 1x3 but 'whole_ts' has is [1:2150, 1:1928, 1:2426].

%% loading TACS

subj = {'subj01' 'subj02' 'subj03'}
trs = [2.8, 2.2, 2.2];
whole_ts = [C{1:2150, 1:1928, 1:2426}];
for k = 1:3
        ts{k} = load(fullfile(D,subj{k},'ts','thalamus_ts.txt'));
        whole_ts = 1:length(ts{k})
        t_mr_1 = (1:length(whole_ts))*(trs(k)/60)
        subplot(3,1,k), hold on
        plot(t_mr_1, ts{k})
        
end
   

Any help would be much appreciated!

[Allen]

Not sure why you are placing a 'C' into your code. Since C does not appear to be a predefined variable, MATLAB assumes that it is a function. However based on the naming, I doubt that you are actually trying to call a function and that it is has been placed in your code by mistake.

whole_ts = [C{1:2150, 1:1928, 1:2426}]; % Invalid syntax and also no need to use [] outside of {}.

Using [] in MATLAB creates a numeric array, {} creates a cell array. Each element in a numeric array need to be a scalar value, where elements in a cell array can consist of various types of data and not just numerical scalars. In the example below there are three elements in the cell array whole_ts, where each elements consists of a numerical array and more specifically for this instance numerical vector arrays.

whole_ts = {1:2150, 1:1928, 1:2426};

Also, can you provide an explanation to why you are defining whole_ts outside of the for-loop and again each iteration of your loop? Does your code perform as expected when removing the usage of whole_ts before your for-loop?

whole_ts = [C{1:2150, 1:1928, 1:2426}]; % Tries to initially define whole_ts but with an error
for k = 1:3
    ts{k} = load(fullfile(D,subj{k},'ts','thalamus_ts.txt'));
    whole_ts = 1:length(ts{k}) % Re-defines whole_ts each loop iteration
    t_mr_1 = (1:length(whole_ts))*(trs(k)/60)
    subplot(3,1,k), hold on
    plot(t_mr_1, ts{k})
end

[Walter Roberson]

whole_ts = [C{1:2150, 1:1928, 1:2426}]; % Invalid syntax and also no need to use [] outside of {}.

That is not invalid syntax, and the [] is meaningful. C{range,range,range} is cell expansion if C is a cell; cell expansion is creating a comma-separated list, and then the [] is horzcat() of that comma-separated list. Entirely valid code (though it is not clear that it is what the user was wanting to do.)

[nines]

Hello!

Thank you both!

Walter, I am trying to time time intervals that I then can use to plot out the time verses the timeseries.

See below:

This is what used to work if all subjects have the same x axis (see line %2) where I am plotting the thalamus timeseries against the time (whole_ts). I am doing this because I have to convert a the time because I have to take in account the TRs (sampling rate for neuroimaging) and then convert to minutes.

subj = {'sub01' 'sub02'}   %1
whole_ts = 1:2150                                         %2

for k = 1:3
        ts{k} = load(fullfile(D,subj{k},'ts','thalamus_ts.txt'));
    
        whole_ts = 1:length(ts{k})
        t_mr_1 = (1:length(whole_ts))*(trs(k)/60)
        subplot(3,1,k), hold on
        plot(t_mr_1, ts{k})
 
end

I am trying to make a script where I can put in different values for the x axis for different subjects, e.g. if I have three subjects with the corresponding time values:

1:2150, 1:1928, 1:2426

Hopefully that's more clear, and thanks a ton!

[nines]

So I thought I fixed the problem, but actually I didn't.

Any help would be much appreciated

[nines]

now i i have gotten the top part to work, but I am having a hard time plotting the variables now:

for k = 1:length(subj)
        ts{k} = load(fullfile(D,subj{k},'ts','thalamus_ts.txt'));
    
        whole_ts{k} = 1:length(ts{k})
        t_mr_1{k} = ((1:length(whole_ts))*(trs(k)/60))
        subplot(3,1,k), hold on
        plot(t_mr_1{k},ts{k}) %now i cannot plot this as a graph
  
end

[Walter Roberson]

What would be the difference compared to your existing code where you compute (1:length(whole_ts)) ?

[nines]

I don't think anything really, except that for some reason this:

whole_ts = [C{1:2150, 1:1928, 1:2426}];

is telling me that indexing is not supported with the C function.

do you have any suggestions on how to plot indexes:

plot(t_mr_1{k},ts{k}) %now i cannot plot this as a graph

I am getting an error that says the the vectors are not the same length, although they are.

[Walter Roberson]

Where did you get the C part from? Why are you expecting it to be there?

whole_ts = {1:2150, 1:1928, 1:2426};

Your file looks like it might be a 2D array instead of a vector. If so then you have to be very careful when you use length() as length(X) is defined as:

temporary = size(X);
if any(temporary == 0)
    length is 0
else
    length is max(temporary)
end

That is, length is 0 if any dimension is 0, and otherwise length is the largest dimension, no matter which one it is.

However, when you ask to

plot(t_mr_1{k},ts{k})

then if ts{k} is a 2D array, then MATLAB would try to plot every column as a separate line, and the length of t_mr_1{k} would have to be the same as the number of rows in ts{k} .