train_models.py

import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.linear_model import LogisticRegression
from sklearn.svm import LinearSVC
from sklearn.metrics import classification_report
import joblib

df = pd.read_csv('cleaned_data.csv')

X = df['cleaned_text']
y = df['transaction_type']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

vectorizer = TfidfVectorizer()
X_train_vec = vectorizer.fit_transform(X_train)
X_test_vec = vectorizer.transform(X_test)

models = {
    "Naive Bayes": MultinomialNB(),
    "Logistic Regression": LogisticRegression(max_iter=1000),
    "SVM": LinearSVC()
}

best_model = None
best_f1 = 0

for name, model in models.items():
    model.fit(X_train_vec, y_train)
    preds = model.predict(X_test_vec)
    report = classification_report(y_test, preds, output_dict=True)
    print(f"\n{name}\n", classification_report(y_test, preds))

    if report['weighted avg']['f1-score'] > best_f1:
        best_f1 = report['weighted avg']['f1-score']
        best_model = model

# Save best model and vectorizer
joblib.dump(best_model, 'final_model.pkl')
joblib.dump(vectorizer, 'tfidf_vectorizer.pkl')