🌐 Website · 📚 Documentation
Interpolation, splines and polynomials in pure Go (CGO=0), validated against SciPy and NumPy.
interp provides the interpolators of scipy.interpolate (1-D and N-D,
cubic, PCHIP, Akima, B-splines, FITPACK smoothing splines) and poly the
polynomial tools of NumPy (polyfit, roots, the numpy.polynomial bases).
Both run SciPy and NumPy as oracles and compare to them value by value.
Chebyshev and AAA rational approximation are planned.
s, _ := interp.NewCubicSpline(x, y, "not-a-knot")
v := s.At(2.5)
fit, _ := poly.Polyfit(x, y, 3, nil)
r, _ := poly.Roots(fit.Coef)| Repo | What it is | Status |
|---|---|---|
| interp | interpolation, splines and smoothing: interp1d, cubic, PCHIP, Akima, B-splines, FITPACK smoothing splines, regular-grid and tensor-product interpolation |
released v0.1.0 |
| poly | polynomials: evaluation, arithmetic, least-squares fits, roots, and the Chebyshev, Legendre, Hermite, HermiteE and Laguerre bases | released v0.1.0 |
| docs | the documentation site (Hugo + Hextra), served at /docs/ | |
| go-approx.github.io | the landing page (Hugo) | |
| brand | logo, icons and social images |