Is there an efficient way to find the indices of the closest matching values in a large matrix, to a large lookup vector?

I have 2 large lookup tables (Q_lut , gamma_lut), with each value in them corresponding to the voltages of two voltage vectors (E1_interp for col, E2_interp for row) respectively. I have acquired raw voltage data pairs in large matrices (E1_inst, E2_inst), which I can use to find the associated Q and gamma based the index of the closest matching value in the voltage vectors.

The issue is, that the code below is very slow, taking roughly 1500 s for this small dataset. The only way I could get the min function to work, was to check each value in E1_inst, E2_inst individually. These are 400000x9 matrices, checking through a 25000x1 vector on every iteration so it takes a long time. I'm sorry if similar questions have been asked before, but does anyone know if there is a more efficient way to do this?

Thanks :)

tic
ind_e1 = zeros(size(E1_inst,1), size(E1_inst,2));
ind_e2 = zeros(size(E1_inst,1), size(E1_inst,2));
Q = zeros(size(E1_inst,1), size(E1_inst,2));
gamma = zeros(size(E1_inst,1), size(E1_inst,2));

for i=1:size(E1_inst,1)
    for j=1:size(E1_inst,2)
        
        [~, ind_e1(i,j)] = min(abs(E1_interp - E1_inst(i,j)));
        [~, ind_e2(i,j)] = min(abs(E2_interp - E2_inst(i,j)));
        Q(i,j) = Q_lut(ind_e2(i,j), ind_e1(i,j));
        gamma(i,j) = Pitch_lut(ind_e2(i,j), ind_e1(i,j));

    end
end
toc

Unrecognized function or variable 'E1_inst'.

Answer 1

[Matt J]

[E1,is]=sort(E1_interp);
[E2,js]=sort(Es_interp);
Q_lut=Q_lut(js,is);
Pitch_lut=Pitch_lut(js,is);

[Q,gamma]=deal(zeros(size(E1_inst,1), size(E1_inst,2)));

Q(:)=interp2(E1,E2,Q_lut,E1_inst(:),E2_inst(:),'nearest');
gamma(:)=interp2(E1,E2,Pitch_lut,E1_inst(:),E2_inst(:),'nearest');

[Daniel Burnett]

Thank you very much, almost 500 times faster for the example I tested!!