Unable to perform assignment because the indices on the left side are not compatible with the size of the right side.

Dear community, help with this.

Voc=Clean(:,7);
a=Voc(1:3);
j=1;k=9;
t=length(Voc);
for i=1:t;
    v= Voc(j:k,1);
    a=cellfun(@mean,v);
    Voc_p(i,1)=a;
    j=j+9;
    k=k+9;
end

A datatable was convert into a array previously and appears this error:

Unable to perform assignment because the indices on the left side are not compatible with the size of the
right side.
Error in Prueba_1 (line 29)
Voc_p(i,1)=a;

How can I fix this?

Thank u

Answer 1

[Walter Roberson]

v= Voc(j:k,1);

That is creating a non-scalar cell array

a=cellfun(@mean,v);

That is going through each cell member and taking the mean of each. The fact that you did not get an error on this line tells us that each cell member was a scalar or a vector, not empty and not 2 D or more. The end result will be a vector of length k-j+1 = 9.

    Voc_p(i,1)=a;

You are trying to store that vector of length 9 into a single location, Voc_p(i,1)

[Jonathan Bijman]

I see. I put a for loop because I need to get the average of nine measurements (from the first item to the nineth item) and the repeat from the 10th to 19th and successively, that's why I wrote

Voc_p(i,1)=a;

Because in that vector I want to save the averages.

I don't know how to get the average in that way, that's why I wrote

a=cellfun(@mean,v);

How can I fix this?

[Walter Roberson]

We do not know how many values are in each cell entry. If it is not a scalar in each cell, then it becomes unclear what you are taking the average of, or what output size you want. If it is a scalar in each cell, then it becomes unclear why you used a cell instead of a numeric array.

Perhaps

Voc_p = mean( reshape( [Clean{:,7}], 9, [] ) );

with no loop

[Jonathan Bijman]

I try to convert the datatable into a array, but it appers this error:

Error using table2array (line 27)
Unable to concatenate the table variables 'Comment' and 'CellParamArea', because their types are cell and
double.
Error in Prueba_1 (line 18)
Clean=table2array(datatable);

That's why I converted it into a cell. Maybe that's the problem.

Applying the last thing you suggested appears this error:

Error using reshape
Product of known dimensions, 9, not divisible into total number of elements, 289.
Error in Prueba_1 (line 20)
Voc_p = mean( reshape( [Clean{:,7}], 9, [] ) );

I don´t know how to get it =(

Thank u for your help

[Walter Roberson]

You want to divide the data into groups of 9, but you have 289 entries, which is 32 groups of 9 plus one left over. What do you want to do about that one left over?

[Jonathan Bijman]

I want to save in one column, the average of 9 measurements (it would be 32 averages) and then I want to plot the measurements along the time.

Although there is 39 measurements, there was some days when weren´t measuments. (That's why I converted into a datatable, because I want the commands recognize the days where weren´t measurements and continue the plot along the time)

I hope that help u.

[Walter Roberson]

nrow = height(datatable);
lastidx = floor(nrow/9) * 9:

Voc_p = mean( reshape(datatable{1:lastidx,7}, 9, []) );

[Jonathan Bijman]

appears this error:

Error using reshape
Product of known dimensions, 9, not divisible into total number of elements, 289.
Error in Prueba_1 (line 23)
Voc_p = mean( reshape(datatable{1:lastidx,7}, 9, []) );

[Walter Roberson]

That makes no sense after the way I calculated lastidx with floor. Please show the value of nrow, and the value of lastidx, and size(datatable{1:lastidx,7})

[Jonathan Bijman]

nrow = 289

lastidx = 289

size(datatable {1:lastidx,7}) = 289 1

[Walter Roberson]

Please post the line of code you use to calculate lastidx

[Jonathan Bijman]

clc
clear all
datadir = 'C:\Users\Jonat\Desktop\Clean';
dinfo = dir( fullfile(datadir, '*.xlsx') );
filenames = fullfile( datadir, {dinfo.name} );
numfiles = length(filenames);
datatable = [];
for K = 1 : numfiles
    thisfile = filenames{K};
    thistable = readtable(thisfile);
    if K == 1
        datatable = thistable;
    else
        datatable = [datatable; thistable];
    end
end

Clean=table2cell(datatable);
Voc=Clean(:,7);
a=Voc(1:3);
nrow = height(datatable ); 
lastidx = floor((nrow/9) * 9);
Voc_p = mean( reshape(datatable{1:lastidx,7}, 9, []) );

that's the whole script I'm using

[Walter Roberson]

You moved the *9 to inside the floor() when it needs to be outside.

lastidx = floor(nrow/9) * 9;

[Jonathan Bijman]

It works!

Thank U so much Walter for all the troubles I caused u.

I hope can I count for your help in the future, for this amateur like me.

Thank u so much =)