GSLS.gradient_approximation

GSLS.gradient_approximation(n, x, x_value, directions, sample_set_x, sample_set_y)

Construct gradient approximation from given sample.

Parameters

• n (int) – Dimension of the problem.

• x (ndarray) – Point around which the sample set was constructed.

• x_value (float) – Objective function value at x.

• directions (ndarray) – Directions of the sample points wrt the central point x, as a 2D array.

• sample_set_x (ndarray) – x-coordinates of the sample set, one point per row, as a 2D array.

• sample_set_y (ndarray) – Objective function values of the points in sample_set_x, as a 1D array.

Return type

ndarray

Returns

Gradient approximation at x, as a 1D array.