Hello,
As per my understanding, the line of code mentioned above will calculate the gradient of the sum of elements in the array `y` with respect to the input `dlX`.
Thus, the variable `dy` will have the same dimension as the input `dlX`.
You may use the `quiver` function to plot the gradients calculated using the `dlgradient` function.
The `quiver (X, Y, U, V) ` function in MATLAB plots arrows with directional components U and V at the Cartesian coordinates specified by X and Y. Each arrow represents a vector starting from the corresponding point (X, Y) and extending horizontally by U and vertically by V. By default, the function scales the arrow lengths to prevent overlapping.
Rosenbrock’s function is a standard test function for optimization. You may use it for demonstration purpose and enabled the higher derivative option. Then, compute the gradient using automatic differentiation, before plotting it using the `quiver` function.
You may find the code in the attached live script file (.mlx file), for your reference.
References:
Hope this helps.
