Wrong left and right assignment error

Hi there,

I have ellipsoids of different principal radii a b c and within these I attempt to quantify the number of smaller spheres with constant radii that I can fit into the ellipsoids and assess how the number of spheres changes with increasing ellipsoid axes.

The relationship of the counts and the size of the ellipsoids is as follows:

atom counts = 4/3*pi*a*b*c (2/0.0715)

I am still new to matlab and I am struggling to get a proper expression to assess this.

The code below gives the following error:

Unable to perform assignment because the left and right sides have a different number of elements.

a = [3.4995 3.098 2.737 1.9185 2.1555 2.6865 2.0295 2.3715 1.111 2.1575 1.93 1.6605 1.884 1.9795 3.0115 1.2425 2.569 1.8575 1.4275 1.427 2.2905];
b = [2.29 2.807 2.1635 1.6945 1.1335 2.0845 1.7195 2.013 0.9385 1.68 1.7545 0.9175 1.0685 1.029 2.392 0.986 2.1365 1.576 1.0705 1.104 1.882];
c =[2.035 1.295 2.035 1.295 1.295 1.295 1.295 1.295 0.74 1.48 0.925 0.925 1.665 1.11 1.665 0.74 2.035 1.295 0.74 0.555 1.665];
f_pi= 4.1888;
ucellatoms= 2/0.0715;

for n= 1:1:100
    l(n) = n.*(a.*b.*c);
    Mean_diam(n) =(n.*(a+b+c)./3);
    atm_count = ucellatoms.*f_pi.*l;
    plot (atm_count,l,'o')
    nabc=l';
  mean_d = Mean_diam';
   atcounts=atm_count';
   Ex = [nabc,atcounts];
   plot(mean_d,atcounts, '*')
end 

Answer 1

[Jakob B. Nielsen]

Your problem is that l(n) wants to place a single value into the array l in index n. But your a, b and c which you multiply with each other, are 1x21 arrays - as such the product of these is itself a 1x21 array. As such, you need to specify l (and Mean_diam as well) to be arrays themselves. The below will run, but I dont know if it will produce what you want it to :)

a = [3.4995 3.098 2.737 1.9185 2.1555 2.6865 2.0295 2.3715 1.111 2.1575 1.93 1.6605 1.884 1.9795 3.0115 1.2425 2.569 1.8575 1.4275 1.427 2.2905];
b = [2.29 2.807 2.1635 1.6945 1.1335 2.0845 1.7195 2.013 0.9385 1.68 1.7545 0.9175 1.0685 1.029 2.392 0.986 2.1365 1.576 1.0705 1.104 1.882];
c =[2.035 1.295 2.035 1.295 1.295 1.295 1.295 1.295 0.74 1.48 0.925 0.925 1.665 1.11 1.665 0.74 2.035 1.295 0.74 0.555 1.665];
f_pi= 4.1888;
ucellatoms= 2/0.0715;
for n= 1:1:100
    l(:,n) = n.*(a.*b.*c);
    Mean_diam(:,n) =(n.*(a+b+c)./3); 
    atm_count = ucellatoms.*f_pi.*l;
    plot (atm_count,l,'o')
    nabc=l';
  mean_d = Mean_diam';
   atcounts=atm_count';
   Ex = [nabc,atcounts];
   plot(mean_d,atcounts, '*')
end 

[Arthur Moya]

Hi @Jakob B. Nielsen,

Thank you for your response,

I have attempted to run the code but I run into another error when I do.

Kind regards,

Arthur Moya

Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is
1-by-21.

Error in Ellipse_Atom_Count (line 7)
    l(:,n) = n.*(a.*b.*c);
 

[Jakob B. Nielsen]

Have you cleared The old l? If Its still a 1x1 it Will face The same problem. clear l in The command Window / top of The code and try again? The piece i posted ran fine for me.

[Arthur Moya]

It works well thank you.

I am curious to understand what the l(:,n) syntex means.

Your response will be highly appreciated.

Kind regards

Arthur Moya

[Jakob B. Nielsen]

Well the first entry is row number, and l(:,1) = x simply means all rows in The first column. If you put a 2 row vector as x, The resulting l Will be a 2 by 1 array, and if your x is insteaf a 5000 row vector, your l Will be 5000 by 1. I hope that makes sense.

Answer 2

[Steven Lord]

for n= 1:1:100
    l(n) = n.*(a.*b.*c);

When you execute that second line of code, n is a scalar (size 1-by-1.) The variables a, b, and c are 1-by-21 vectors. The result of evaluating n.*(a.*b.*c) is also 1-by-21. You're trying to stuff a 1-by-21 vector into one element of the variable l. That's not going to work; it doesn't fit.

What size do you want / expect the variables l and Mean_diam to be when this code finishes?

[Arthur Moya]

Hi @ Steven Lord,

Thank you for your response.

I expect 'l' and 'Mean_diam' to be 100x1 vectors for every (a.*b.*c) to produce 21 curves with a 100 points.

My goal is to proportionally expand a b c using n from 1 to a 100 for every combination of a b c.

I didn't know how to best express the different values of a b and c such that an elementwise operation is done.

(I hope this makes some sense)

Kind regards,

Arthur Moya