Implementation:Scikit learn Scikit learn FetchOlivettiFaces

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

Overview

Concrete tool for fetching and loading the Olivetti faces dataset for face recognition tasks, provided by scikit-learn.

Description

The fetch_olivetti_faces function downloads and caches the modified Olivetti faces dataset. The dataset contains 400 grayscale face images of 40 subjects (10 images per subject), each of size 64x64 pixels. The images were taken at AT&T Laboratories Cambridge and are provided in MATLAB format from Sam Roweis's personal web page.

Usage

Use this function when you need a face recognition dataset for evaluating classification, clustering, or dimensionality reduction algorithms. It is commonly used in tutorials for PCA, NMF, and face recognition examples.

Code Reference

Source Location

• Repository: scikit-learn
• File: sklearn/datasets/_olivetti_faces.py

Signature

@validate_params(...)
def fetch_olivetti_faces(
    *,
    data_home=None,
    shuffle=False,
    random_state=0,
    download_if_missing=True,
    return_X_y=False,
    n_retries=3,
    delay=1.0,
):

Import

from sklearn.datasets import fetch_olivetti_faces

I/O Contract

Inputs

Name	Type	Required	Description
data_home	str, PathLike or None	No	Custom directory for caching (default None)
shuffle	bool	No	Whether to shuffle the dataset (default False)
random_state	int, RandomState or None	No	Random seed for shuffling (default 0)
download_if_missing	bool	No	If True, download data if not cached (default True)
return_X_y	bool	No	If True, return (data, target) instead of Bunch (default False)
n_retries	int	No	Number of download retries (default 3)
delay	float	No	Delay between retries in seconds (default 1.0)

Outputs

Name	Type	Description
dataset	Bunch	Dictionary-like object with data (400x4096), target, images (400x64x64), and DESCR
(X, y)	tuple of ndarray	Feature matrix and target array when return_X_y=True

Usage Examples

Basic Usage

from sklearn.datasets import fetch_olivetti_faces

faces = fetch_olivetti_faces()
print(faces.data.shape)    # (400, 4096)
print(faces.images.shape)  # (400, 64, 64)
print(faces.target.shape)  # (400,)

X, y = fetch_olivetti_faces(return_X_y=True, shuffle=True, random_state=42)

Related Pages

• Principle:Scikit_learn_Scikit_learn_Dataset_Loading