SMTorg/smt

Implements gradient-enhanced surrogate modeling with first and second-order derivative support throughout the pipeline—from training data to predictions. Core models include kriging with partial-least squares reduction and energy-minimizing spline interpolation, alongside classical methods like Gaussian processes and radial basis functions. Integrates with NumPy/SciPy ecosystem and includes sampling techniques and benchmark functions for design optimization workflows.

861 stars. Actively maintained with 8 commits in the last 30 days. Available on PyPI.