Update app.py

app.py

@@ -1,24 +1,25 @@
 import streamlit as st
 import pandas as pd
-import joblib
 import re
 import string
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.naive_bayes import MultinomialNB
+from sklearn.model_selection import train_test_split
 
 # Page config
 st.set_page_config(page_title="SMS Spam Detector", layout="centered")
 st.title("📩 SMS Spam Detection App")
 st.markdown("🔍 Enter a message below to check if it's **Spam** or **Not Spam (Ham)**")
 
-# --- Load CSV for reference or stats ---
-csv_path = "https://huggingface.co/spaces/MLDeveloper/Spam_SMS_Detection/resolve/main/spam.csv"
+# --- Load dataset ---
+csv_url = "https://huggingface.co/spaces/MLDeveloper/Spam_SMS_Detection/resolve/main/spam.csv"
 try:
-    df = pd.read_csv(csv_path)
+    df = pd.read_csv(csv_url, encoding='latin-1')[['v1', 'v2']]
+    df.columns = ['label', 'message']
+    df['label'] = df['label'].map({'ham': 0, 'spam': 1})
 except Exception as e:
-    st.error(f"Error loading dataset: {e}")
-
-# --- Load trained model & vectorizer ---
-model = joblib.load("model/spam_model.pkl")  # ✅ your trained model
-vectorizer = joblib.load("model/tfidf_vectorizer.pkl")  # ✅ your TF-IDF vectorizer
+    st.error(f"Failed to load CSV: {e}")
+    st.stop()
 
 # --- Text Cleaning Function ---
 def clean_text(text):
@@ -30,6 +31,18 @@ def clean_text(text):
     text = text.translate(str.maketrans('', '', string.punctuation))
     return text.strip()
 
+df['cleaned'] = df['message'].apply(clean_text)
+
+# --- Train model ---
+X = df['cleaned']
+y = df['label']
+
+vectorizer = TfidfVectorizer()
+X_vec = vectorizer.fit_transform(X)
+
+model = MultinomialNB()
+model.fit(X_vec, y)
+
 # --- Prediction Function ---
 def predict_spam(message):
     cleaned = clean_text(message)
@@ -37,7 +50,7 @@ def predict_spam(message):
     prediction = model.predict(vector)
     return "Spam" if prediction[0] == 1 else "Not Spam"
 
-# --- UI for prediction ---
+# --- UI ---
 user_input = st.text_area("✉️ Enter your SMS message here:")
 
 if st.button("Check Message"):
@@ -50,9 +63,9 @@ if st.button("Check Message"):
         else:
             st.success("✅ This message is classified as **NOT SPAM (HAM)**.")
 
-# Optional: Show dataset preview
+# Optional: View CSV
 with st.expander("📄 View sample dataset (CSV)"):
-    st.dataframe(df.head())
+    st.dataframe(df[['label', 'message']].head())
 
 st.markdown("---")
-st.markdown("🔒 **Note**: This is a demo model and not intended for production use without proper testing.")
+st.markdown("🔒 **Note**: Model is trained in real-time from CSV and not saved for reuse. Ideal for demo purposes.**")