Implementation:Scikit learn Scikit learn FetchKddcup99

Knowledge Sources	Scikit_learn Scikit-learn Docs
Domains	Data Loading, Anomaly Detection
Last Updated	2026-02-08 15:00 GMT

Overview

Concrete tool for fetching the KDD Cup 99 intrusion detection dataset provided by scikit-learn.

Description

This module implements the fetch_kddcup99 function that downloads and loads the KDD Cup 1999 dataset, a classic benchmark for anomaly detection and network intrusion detection. The dataset contains network connection records with labels indicating normal connections or specific attack types. The module supports loading various subsets (SA, SF, http, smtp), the full dataset or 10% sample, and can return data as NumPy arrays or pandas DataFrames.

Usage

Use this function to load the KDD Cup 99 dataset for evaluating anomaly detection, network intrusion detection, or outlier detection algorithms.

Code Reference

Source Location

• Repository: scikit-learn
• File: sklearn/datasets/_kddcup99.py

Signature

@validate_params(...)
def fetch_kddcup99(
    *,
    subset=None,
    data_home=None,
    shuffle=False,
    random_state=None,
    percent10=True,
    download_if_missing=True,
    return_X_y=False,
    as_frame=False,
    n_retries=3,
    delay=1.0,
)

Import

from sklearn.datasets import fetch_kddcup99

I/O Contract

Inputs

Name	Type	Required	Description
subset	str or None	No	Subset to load: 'SA', 'SF', 'http', 'smtp', or None for full dataset
data_home	str or PathLike or None	No	Custom directory for data storage
shuffle	bool	No	Whether to shuffle the dataset (default: False)
random_state	int or None	No	Random state for reproducible shuffling
percent10	bool	No	Whether to load 10% subset (default: True)
return_X_y	bool	No	If True, return (data, target) tuple (default: False)
as_frame	bool	No	If True, return data as pandas DataFrame (default: False)

Outputs

Name	Type	Description
data	Bunch	Dictionary-like object with data, target, feature_names, target_names, DESCR
(X, y)	tuple	Returned when return_X_y=True; feature matrix and target array

Usage Examples

Basic Usage

from sklearn.datasets import fetch_kddcup99

# Load the http subset
data = fetch_kddcup99(subset='http', percent10=True)
print("Shape:", data.data.shape)
print("Target classes:", set(data.target))

Related Pages

• Principle:Scikit_learn_Scikit_learn_Dataset_Loading