Anwaree's picture
train_spam_model.py
483e6f5 verified
Raw History Blame Contribute Delete
2.03 kB
import gradio as gr
import joblib
import re
import pandas as pd
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
import nltk
# -----------------------------
# 1️⃣ Prétraitement
# -----------------------------
nltk.download('stopwords')
stop_words = set(stopwords.words('english'))
stemmer = PorterStemmer()
def preprocess_message(text):
"""
Prétraitement générique pour messages inconnus (nouveaux messages à prédire)
: garde ponctuation utile pour spam
"""
if pd.isna(text):
return ""
text = text.lower()
text = re.sub(r'http\S+|www\S+', '', text)
text = re.sub(r'\S+@\S+', '', text)
text = re.sub(r'\+?\d[\d -]{8,}\d', '', text)
text = re.sub(r'\d+', '', text)
# garder ponctuation typique spam
text = re.sub(r'[^a-z\s!/+>]', '', text)
words = [stemmer.stem(word) for word in text.split() if word not in stop_words]
return " ".join(words)
# -----------------------------
# 2️⃣ Chargement du modèle
# -----------------------------
model = joblib.load("spam_model.pkl")
vectorizer = joblib.load("tfidf_vectorizer.pkl")
# -----------------------------
# 3️⃣ Fonction de prédiction
# -----------------------------
def predict_message(message):
cleaned = preprocess_message(message)
X = vectorizer.transform([cleaned])
prediction = model.predict(X)[0]
probability = model.predict_proba(X)[0][1] if hasattr(model, 'predict_proba') else None
return {
"Message": message,
"Prediction": prediction,
"Spam Probability": round(float(probability), 4) if probability is not None else None
}
# -----------------------------
# 4️⃣ Interface Gradio
# -----------------------------
iface = gr.Interface(
fn=predict_message,
inputs=gr.Textbox(lines=3, placeholder="Entrez votre message ici..."),
outputs="json",
title="📩 Spam Detector",
description="Un modèle ML qui détecte si un message est SPAM ou HAM."
)
if __name__ == "__main__":
iface.launch()