Update app.py

app.py

@@ -48,6 +48,22 @@ def rnn_predict_message(input_text):
         else:
             return "Not spam"
 
+# Load the saved LSTM model
+lstm_smsspam_model=tf.keras.models.load_model('lstm_smsspam_model.h5')
+# Load the saved tokenizer
+with open('lstm_smsspam_tokenizer.pickle', 'rb') as handle:
+    lstm_smsspam_tokenizer = pickle.load(handle)
+
+def lstm_predict_message(message):
+    maxlen=50
+    sequence = lstm_smsspam_tokenizer.texts_to_sequences([message])
+    sequence = tf.keras.preprocessing.sequence.pad_sequences(sequence, padding='post', maxlen=maxlen)
+    prediction = lstm_smsspam_model.predict(sequence)[0, 0]
+    if prediction > 0.5:
+        return 'Spam'
+    else:
+        return 'Not spam'
+
 
 
 # Main function for Streamlit app
@@ -90,6 +106,17 @@ def main():
                     st.write(f"The message is classified as: {prediction_result}")
                 else:
                     st.write("Please enter some text for prediction")
+
+        elif model == "LSTM":
+            st.subheader("SMS Spam Detection")
+            user_input = st.text_area("Enter a message to classify as 'Spam' or 'Not spam': ")
+                
+            if st.button("Predict"):
+                if user_input:
+                    prediction_result = lstm_predict_message(user_input)
+                    st.write(f"The message is classified as: {prediction_result}")
+                else:
+                    st.write("Please enter some text for prediction")
                 
     
     elif task == "Movie Sentiment Analysis":