scipy.interpolate.NdBSpline.

design_matrix

classmethod NdBSpline.design_matrix(xvals, t, k, extrapolate=True)

Construct the design matrix as a CSR format sparse array.

Parameters:

xvalsndarray, shape(npts, ndim)

Data points. xvals[j, :] gives the j-th data point as an ndim-dimensional array.

ttuple of 1D ndarrays, length-ndim

Knot vectors in directions 1, 2, … ndim,

kint

B-spline degree.

extrapolatebool, optional

Whether to extrapolate out-of-bounds values of raise a ValueError

Returns:

design_matrixa CSR array

Each row of the design matrix corresponds to a value in xvals and contains values of b-spline basis elements which are non-zero at this value.