How do I calculate partial derivatives numerically for a multi dimensional function?

Hello,

I have a 6 dimentional function, which is a trained Gaussian process regression model obtained using the Regression Learner Matlab app.

I want to calculate the particual detivatives numerically with respect to each of the dimentions.

How can I do this, when the function is of the form trainedModel.predictFcn(x), where x is the 6 dimentional input vector?

Many thanks!

p.s. the reason I want to do this, is to use the Newton Raphson method to solve the equation f(x)=Y, please see my other question here if you can kindly give any further help:

https://uk.mathworks.com/matlabcentral/answers/526197-how-can-i-find-the-solution-to-a-multi-dimensional-struct-function-in-matlab