train_logistic_model.py

import pandas as pd
import re
import os
import joblib
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split

# Load data
df = pd.read_csv("data/spam.csv", encoding='latin-1')
df = df[['v1', 'v2']]
df.columns = ['label', 'message']
df.drop_duplicates(inplace=True)
df['message'] = df['message'].fillna("")
df['label'] = df['label'].fillna(df['label'].mode()[0])

# Clean text
def clean_text(text):
    text = text.lower()
    text = re.sub(r'\W', ' ', text)
    text = re.sub(r'\s+[a-zA-Z]\s+', ' ', text)
    text = re.sub(r'\s+', ' ', text)
    return text.strip()

df['message'] = df['message'].apply(clean_text)
df['label'] = df['label'].map({'ham': 0, 'spam': 1})

# Train-test split
X_train, X_test, y_train, y_test = train_test_split(
    df['message'], df['label'], test_size=0.2, random_state=42
)

# Vectorize
vectorizer = TfidfVectorizer(max_features=5000)
X_train_vec = vectorizer.fit_transform(X_train)
X_test_vec = vectorizer.transform(X_test)

# Train model
model = LogisticRegression()
model.fit(X_train_vec, y_train)

# Save model
os.makedirs("Models", exist_ok=True)
joblib.dump(model, "Models/logistic_model.pkl")
joblib.dump(vectorizer, "Models/logvectorizer.pkl")

print("✅ Logistic model trained & saved successfully!")