A question about the Jacobian matrix

[RonghaiHe]

Hello, Mr. Urban, thank you for your open source at first. I have a question about the Jacobian matrix in MLPnP. Does the Jacobian matrix come from the partial derivative of the eq(11) in your paper? If so, I'd like to know why it is so complicated. For example, is it $\frac{\partial (err)}{\partial t_1} = r_1$ or $s_1$?

[lturing]

@RonghaiHe hi, do you figure it out? i have the same question as you.

[urbste]

Hey guys, sorry for the late response. When I implemented the algorithm in Matlab I just used the symbolic toolbox to output the jacobians with the ccode("") function. There is probably a more elegant, simple analytical solution for this. I just did not bother to calculate it.