Third for loop runs, second and third loops do not

Hi there,

I have searched all over Matlab Answers for a solution to this problem but can't find one.

I have a function which contains the 3 for loops below. When I run the function, only the third for loop is executed (which I have diagnosed using the disp statements).

Is there a reason why these three loops, in this order, would not all be executed? I have been using Matlab for a number of years now but am completely baffled and confused. When I changed the first for loop to "tt = 1:val_T", that loop ran, so I suspect it has something to do with the indexing? Unfortunately, in many cases in my code the 1st index is special, so I need my loops to run from 2 onwards, so I would love a solution which does not require me to change all of my indexing to start from 1 (I have no idea how I could do such a thing, and have no idea why Matlab would require such a thing!).

I would very much appreciate any help, and am happy to provide the full script if there are no problems with the code below. Thank you so much!

for tt = 2:val_T
    disp('in first tt loop')
    inputN(:, tt) = iWN(:, tt) + (inputN(:, tt-1) - iWN(:, tt))*exp(-dtModel);
end

for tt = 2:val_T-1
    disp('in second tt loop')
    val_R(:, tt) = tanh(val_H(:, tt));
    for channel = 1:C
        if val_inputs(channel, tt) == 1 
            stimuli_epochs(channel) = 1; 
        end
    end
    val_JR = J * val_R(:, tt) + inputN(:, tt);
    for channel = 1:C
        neuronal_inputs(channel, :, tt) = amp_rgc * val_inputs(channel, tt) .* input_weights(channel);
    end
    for channel = 1:C
        val_JR = val_JR + transpose(neuronal_inputs(channel, :, tt));
    end
    disp('val_JR: ')
    disp(val_JR)
    val_H(:, tt+1) = val_H(:, tt) + dtModel * (- val_H(:, tt) + val_JR);
end

for tt = 1:val_T
    disp('in third tt loop')
    % check if there is a data timepoint to match with model timepoint
    if model_speed == 1
        data_index = (tt + model_speed - 1) / model_speed;
        JR_err = val_JR - val_data(:, data_index);
        R_err = val_R(:, tt) - val_data(:, data_index);
        H_err = val_H(:, tt) - val_data(:, data_index);
    elseif mod(tt, model_speed) == 1
        data_index = (tt + model_speed - 1) / model_speed;
        JR_err = val_JR - val_data(:, data_index);
        R_err = val_R(:, tt) - val_data(:, data_index);
        H_err = val_H(:, tt) - val_data(:, data_index);
    end
    JR_mse = JR_mse + mean(JR_err.^2);
    R_mse = R_mse + mean(R_err.^2);
    H_mse = H_mse + mean(H_err.^2);
end

Answer 1

[Torsten]

If val_T were 1, only the third loop was entered. But we don't know the value of val_T.

[jchooch]

OMG. Yes, this was the problem! Thank you so much.