gamfit 0.1.99

Formula-first generalized additive models with a high-performance Rust core

gamfit

Formula-first generalized additive models for Python, backed by a Rust engine.

gamfit fits Gaussian, binomial, Poisson, and Gamma GLMs with smooth terms, random effects, bounded/constrained coefficients, location-scale extensions, survival likelihoods, and flexible/learnable links. Smoothing parameters are selected by REML or LAML. Posterior sampling uses NUTS where supported, and Gaussian Laplace for model classes without an exact NUTS path.

Geometric / manifold smooths handle predictor spaces that wrap, including circles, cylinders, tori, the sphere (intrinsic Wahba / spherical harmonic kernels), and periodic tensor products, with no seams or pole artefacts. The gallery's Möbius-looking example is a 4π-periodic double-cover parameterization of a Möbius embedding, not a twisted Möbius-strip predictor basis:

Docs: https://gamfit.readthedocs.io/

Install

uv add gamfit

Wheels for Linux (x86_64, aarch64), macOS (x86_64, Apple silicon), and Windows. No Rust toolchain required.

30-second tour

import gamfit

train = [
    {"y": 1.2, "x": 0.0},
    {"y": 1.9, "x": 1.0},
    {"y": 3.1, "x": 2.0},
    {"y": 4.5, "x": 3.0},
]

model = gamfit.fit(train, "y ~ s(x)")
print(model.predict([{"x": 1.5}, {"x": 2.5}], interval=0.95))
print(model.summary())
model.save("model.gam")

pandas, polars, pyarrow, numpy, dict-of-columns, and list-of-records inputs all work without conversion.

What's different

• Three-part penalty structure. Each smooth gets separate penalties for magnitude, gradient, and curvature. Most GAM libraries use one or two.
• Flexible link functions. Spline offsets from a base link (link(type=flexible(probit))), blended mixture links, and SAS / beta-logistic learnable shapes.
• Surface smooths in arbitrary dimension. Thin-plate, Duchon (with triple-operator regularization), and Matérn covariance, with automatic knot placement.
• Geometric / manifold smooths. Cyclic 1-D, cylinder / torus tensor, intrinsic sphere (Wahba + spherical harmonics), Möbius double-cover demo, boundary-conditioned B-splines. Predictor spaces that wrap or close are first-class — no seams, no pole artefacts.
• Adaptive anisotropy. Per-axis spatial anisotropy shrinks or stretches each feature axis independently inside a single joint smooth.
• Composable basis/kernel. Combine a spline kernel with a length-scale behaviour (e.g. Duchon kernel with Matérn-style global κ).
• Marginal-slope models. Separate baseline risk from a calibrated score's effect, for both Bernoulli and survival outcomes.
• Posterior sampling. model.sample(...) runs NUTS where supported, and Gaussian Laplace for model classes without an exact NUTS path, behind one API.

Highlights from the API

import gamfit
from gamfit.sklearn import GAMRegressor, GAMClassifier

# Validate before you fit
gamfit.validate_formula(train, "y ~ s(x) + group(site)")

# Posterior sampling and mean bands
posterior = model.sample(train, seed=42)
bands = posterior.predict(test, level=0.95)

# Survival
gamfit.fit(df,
    "Surv(entry, exit, event) ~ s(age) + bmi + timewiggle(internal_knots=6)",
    survival_likelihood="transformation",
    baseline_target="weibull",
)

# scikit-learn
est = GAMRegressor(formula="y ~ s(x)")
est.fit(X, y)

# Diagnose, plot, report
model.diagnose(train).metrics
model.plot(train, x="x", kind="prediction")
model.report("report.html")

Public API

Symbol	Purpose
gamfit.fit(data, formula, **kwargs)	Fit a model.
gamfit.load(path) / gamfit.loads(bytes)	Reload a saved model.
gamfit.validate_formula(data, formula, ...)	Type-check a formula without fitting.
gamfit.build_info()	Native extension build metadata.
gamfit.explain_error(exc)	Human-readable hint for a gamfit exception.
gamfit.Model	Fitted-model handle: predict, summary, check, diagnose, plot, report, sample, save.
gamfit.SurvivalPrediction	Per-row hazard / survival surface; on-demand evaluation.
gamfit.SamplingConfig, PosteriorSamples, PosteriorPredictive	NUTS / posterior interface.
gamfit.sklearn.GAMRegressor / GAMClassifier	scikit-learn estimators.

Full reference at https://gamfit.readthedocs.io/en/latest/api-reference/.

Optional extras

uv add "gamfit[pandas]"     # pandas + pyarrow input/output
uv add "gamfit[plot]"       # matplotlib-based plotting
uv add "gamfit[sklearn]"    # scikit-learn integration
uv add "gamfit[torch]"      # PyTorch bridge
uv add "gamfit[all]"        # everything

License

AGPL-3.0-or-later. See LICENSE.