Hybrid Grid Sampler

A hybrid constraint-based sampler combining LHS and grid sampling.

Overview

  • Uses Latin Hypercube Sampling (LHS) for continuous features

  • Uses grid-like random sampling for discrete features (integer, binary, categorical)

  • Allows custom constraints

  • Parallel sampling supported

API Reference

class mlsampler.engine.hypergrid.HyperGridSampler(config)

Bases: BaseSampler

Grid and LHS-based sampler for even coverage of the feature space.

This sampler generates samples based on feature metadata inferred from training data. Float features are drawn by Latin Hypercube Sampling, every other dtype from a uniform grid.

Parameters:

config (SamplerConfig) – Configuration object containing feature metadata and sampling settings.

Notes

This sampler does not support constraints. set_constraints raises NotImplementedError; use RandomSampler for constraint-based sampling.

scipy is required. Its ImportError is deferred until sampling starts.

sample(n_samples: int) → Any
Parameters:

n_samples (int) – Total number of samples to generate.

Returns:

Shape (n_samples, n_features), in the input’s column order. The same type setup was given: an object array for an array, or a pandas/polars DataFrame carrying the input’s column names and dtypes.

Return type:

np.ndarray or DataFrame

Notes

  • Float features are scaled to their respective [low, high] ranges.

  • Integer features are sampled from the inclusive range [low, high].

  • Categorical features are sampled from the provided category list.

  • Constant features return the same value for all samples.