Implementation:Scikit learn Scikit learn BenchRandomProjections

Knowledge Sources	Scikit_learn Scikit-learn Docs
Domains	Machine Learning, Benchmarking
Last Updated	2026-02-08 15:00 GMT

Overview

Concrete tool for benchmarking random projection transformers provided by scikit-learn.

Description

This benchmark script evaluates the performance of GaussianRandomProjection and SparseRandomProjection transformers. It measures fit and transform times across varying data sizes and dimensionalities. The script also validates the Johnson-Lindenstrauss lemma by comparing minimum dimension estimates with empirical results from both dense and sparse random projection implementations.

Usage

Use this benchmark to compare the speed of Gaussian vs. Sparse random projections for dimensionality reduction, and to validate the Johnson-Lindenstrauss bounds for your data characteristics.

Code Reference

Source Location

• Repository: scikit-learn
• File: benchmarks/bench_random_projections.py

Signature

def type_auto_or_float(val)
def type_auto_or_int(val)
def compute_time(t_start, delta)
def bench_scikit_transformer(X, transformer)

from sklearn.random_projection import (
    GaussianRandomProjection,
    SparseRandomProjection,
    johnson_lindenstrauss_min_dim,
)

Import

from sklearn.random_projection import GaussianRandomProjection, SparseRandomProjection
from sklearn.random_projection import johnson_lindenstrauss_min_dim

I/O Contract

Inputs

Name	Type	Required	Description
X	array-like	Yes	Input data matrix to be projected
transformer	object	Yes	Random projection transformer instance
--n-times	int	No	Number of benchmark repetitions
--n-features	int	No	Number of features in input data
--n-components	str/int	No	Target dimensionality or 'auto'

Outputs

Name	Type	Description
time_to_fit	float	Time to fit the transformer in seconds
time_to_transform	float	Time to transform the data in seconds
Console output	text	Timing results for each configuration

Usage Examples

Basic Usage

from sklearn.random_projection import SparseRandomProjection
import numpy as np

X = np.random.randn(10000, 500)
transformer = SparseRandomProjection(n_components=100, random_state=42)
X_reduced = transformer.fit_transform(X)
print("Reduced shape:", X_reduced.shape)

Related Pages

• Principle:Scikit_learn_Scikit_learn_Dimensionality_Reduction