pipeline.py

import pandas as pd
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score
from sklearn.model_selection import train_test_split
from sklearn.metrics import precision_recall_curve
import matplotlib.pyplot as plt


def encode_features(df, exclude_columns):
    """
    df: dataframe
    exclude_columns: numerical columns to exclude in the dataframe
    return: return an encoded dataframe
    """
    from sklearn.preprocessing import LabelEncoder

    le = LabelEncoder()

    categorical_features = df.select_dtypes(["object"])

    # Exclude numerical columns + Churn column
    features_to_encode = categorical_features.loc[
        :,
        ~categorical_features.columns.isin(exclude_columns),
    ].copy()

    # Select the columns to encoded
    columns_to_encode = features_to_encode.columns

    for column in columns_to_encode:
        features_to_encode[column] = le.fit_transform(features_to_encode[column])

    return features_to_encode


def concat_encoded_numerical(df_encoded_features, df_numerical_features):
    """
    df_encoded_features: encoded features dataframe
    df_numerical_features: numerical dataframe

    return: contacted dataframe encoded + numerical
    """
    return pd.concat([df_encoded_features, df_numerical_features], axis=1)


def split_data(X, y, test_size=0.2):

    return train_test_split(X, y, test_size=test_size, random_state=42)


def train_model(model, X_train, y_train):
    rf = model
    rf.fit(X_train, y_train)

    return rf


def metric_model(y_test, y_pred):

    accuracy = accuracy_score(y_test, y_pred)
    precision = precision_score(y_test, y_pred, average="binary")
    recall = recall_score(y_test, y_pred, average="binary")
    f1 = f1_score(y_test, y_pred, average="binary")

    return [accuracy, precision, recall, f1]


def display_pr_curve(model, X_test, y_test, title, label):
    y_probs = model.predict_proba(X_test)[:, 1]

    precision, recall, _ = precision_recall_curve(y_test, y_probs)

    plt.figure(figsize=(8, 6))
    plt.plot(recall, precision, marker=".", color="b", label=label)
    plt.title(title)
    plt.xlabel("Recall")
    plt.ylabel("Precision")
    plt.grid(True)
    plt.legend()
    plt.show()