"reshape" does not give column vector entries in output

Running test_modified.m we obtain the vector entries for the residuals , as a 1x8 vector. However, with either

r = reshape([result{:}].',[],1) % column vector
r = reshape([result{:}].',1,[]) % row vector

I obtain no vector output. It says "Brace indexing is not supported for variables of this type."

What is wrong and how can I correct this?

Thanks

[Dyuman Joshi]

type test_modified.m

result = zeros(8,1);

%experiment  1
y1=[2;1];
dt=0.01;
T=1;
k=[5;1];
enzyme(y0,k,dt,T)
[SP,grad]=enzyme(y0,k,dt,T);
results(1:2) = y1-SP;

%experiment 2
y1=[1;1];
dt=0.01;
T=1;
k=[5;1];
enzyme(y0,k,dt,T)
[SP,grad]=enzyme(y0,k,dt,T);
results(3:4) = y1-SP;

%experime 3
y1=[1;0];
dt=0.01;
T=1;
k=[5;1];
enzyme(y1,k,dt,T)
[SP,grad]=enzyme(y0,k,dt,T);
results(5:6) = y1-SP;

%experiment 4
y1=[4;1];
dt=0.01;
T=1;
k=[5;1];
enzyme(y0,k,dt,T)
[SP,grad]=enzyme(y0,k,dt,T);
results(7:8) = y1-SP

r = reshape([result{:}].',[],1) % column vector

r = reshape([result{:}].',1,[]) % row vector

Based on the code above -

1 - You have preallocated the array as result whereas you are assinging the values to results (extra s at the end). Correct the mis-match.

Also, as the error states, you can not use brace indexing (i.e. indexing via curly brackets - {}) for numeric data type. Only regular indexing is allowed for numeric data types i.e. use parenthesis.

2 - y0 is not defined in test_modified.m.

3 - It will be better to store data (that is varying) in arrays and call enzyme() N number of times, instead of calling it individually every time.

[Stephen23]

@Dyuman Joshi: for point 3 see perhaps: https://www.mathworks.com/matlabcentral/answers/2095876-how-to-extract-data-from-a-file-and-form-a-column-vector#answer_1427096

[Sergio]

Thanks @Dyuman Joshi, but y0 is defined in enzyme.m which is called in test_modified.m

[Sergio]

@Stephen23 good point! In this case, it would be different, and not V = y1(:)-SP(:). With that, I get

V =

3.0146

-1.1146