hapc — Highly Adaptive Principal Components

hapc provides fast HAL projection-based regression, classification, cross-validation, average treatment effect (ATE) estimation, and discrete-time survival (logistic hazard) models, with matching R and Python APIs backed by a shared C++ core.

Contents

• Discrete-time survival (family = "logit-hazard")
  • hazard_hapc()
  • Result type
• API reference
  • High-level model fitting
  • Result types
  • Design & kernels
  • Solvers & single-λ fits
  • Cross-validation

Installation

pip install hapc

Quick start

import numpy as np
from hapc import cv_hapc

rng = np.random.default_rng(0)
X = rng.standard_normal((200, 3))
Y = np.sin(np.pi * X[:, 0]) + X[:, 1] + rng.standard_normal(200) * 0.1

cv = cv_hapc(X, Y, max_degree=2, npcs=199, nfolds=5, norm="sv")
print(cv.best_lambda)

For discrete-time survival, see Discrete-time survival (family = "logit-hazard"). The full project README with installation notes for Linux/HPC clusters and the R package lives on GitHub.

Indices

• Index

• Module Index