Skip to content

Trace Ratio

skfeature.function.similarity_based.trace_ratio

Description

Trace Ratio performs supervised feature selection by optimizing the trace ratio criterion, max trace(Sb) / trace(Sw), where Sb and Sw are the between-class and within-class scatter matrices. It supports a fisher or laplacian style for building the scatter matrices.

Usage

import numpy as np
from sklearn.datasets import load_iris

from skfeature.function.similarity_based import trace_ratio

X, y = load_iris(return_X_y=True)

# rank features with the trace ratio criterion
score = trace_ratio.trace_ratio(X, y)

Parameters

  • X: numpy array, shape (n_samples, n_features) — input data
  • y: numpy array, shape (n_samples,) — class labels
  • n_selected_features: int — number of features to select
  • **kwargs: style ("fisher" or "laplacian") and verbose
  • mode: {{"rank", "index"}}, default "rank" — "rank" returns an array of feature indices ordered by importance and aligned with sklearn.feature_selection.SelectKBest; "index" returns the indices of the selected features with the most important one first

Returns

  • score: numpy array, shape (n_features,) — ranking score of every feature, aligned with sklearn.feature_selection.SelectKBest

References

  • Original implementation from the DMML Lab@ASU Feature Selection Repository.