pranee31/emailClassification

1import pandas as pd2import os3from sklearn.model_selection import train_test_split4from sklearn.ensemble import RandomForestClassifier5from sklearn.pipeline import Pipeline6from sklearn.feature_extraction.text import TfidfVectorizer7from sklearn.metrics import classification_report8import joblib9 10def train_classifier(csv_path: str, model_path: str = "saved_model/email_classifier.pkl"):11    df = pd.read_csv(csv_path)12    if 'email' not in df.columns or 'type' not in df.columns:13        raise ValueError("CSV must have 'email' and 'type' columns.")14 15    X = df['email']16    y = df['type']17 18    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)19 20    clf_pipeline = Pipeline([21        ('tfidf', TfidfVectorizer(stop_words='english')),22        ('clf', RandomForestClassifier(n_estimators=100, random_state=42))23    ])24 25    clf_pipeline.fit(X_train, y_train)26 27    preds = clf_pipeline.predict(X_test)28    print("\nClassification Report:\n", classification_report(y_test, preds))29 30    os.makedirs(os.path.dirname(model_path), exist_ok=True)31    joblib.dump(clf_pipeline, model_path)32    print(f"Model saved to {model_path}")