Documentation
Data Preprocessing
Tabular Data
Feature Selection

Module: feature_selection.py

Functions

METHOD_K_FEATUERS is a parameter that defiend in a CONFIG file
from sklearn.feature_selection import VarianceThreshold, SelectKBest, chi2, mutual_info_regression
from modeling.regression.config import METHOD_K_FEATUERS
from utils.utils import update_active_features
 
def remove_low_variance_features(dataframe, threshold=0.0):
    """
    Remove features with low variance.
 
    Parameters:
        dataframe (pd.DataFrame): The dataframe containing the data.
        threshold (float): Features with a variance lower than this threshold will be removed.
 
    Returns:
        pd.DataFrame: DataFrame with selected features.
    """
 
 
def select_k_best_features(dataframe, labels, k=10, method=METHOD_K_FEATUERS):
    """
    Select the K best features based on the specified method.
 
    Parameters:
        dataframe (pd.DataFrame): The dataframe containing the data.
        labels (pd.Series): The target variable.
        k (int): The number of top features to select.
        method (str): The feature selection method ('mutual_info' or 'chi2').
 
    Returns:
        pd.DataFrame: DataFrame with the K best features selected.
    """
EncodeScale