Update app.py

app.py

@@ -6,11 +6,11 @@ from bs4 import BeautifulSoup
 from datetime import date
 
 # load all the models and vectorizer (global vocabulary)
-# Seq_model = load_model("LSTM.h5") # Sequential
+Seq_model = load_model("LSTM.h5") # Sequential
 SVM_model = joblib.load("SVM_Linear_Kernel.joblib") # SVM
 logistic_model = joblib.load("Logistic_Model.joblib") # Logistic
 vectorizer = joblib.load("vectorizer.joblib") # global vocabulary (used for Logistic, SVC)
-# tokenizer = joblib.load("tokenizer.joblib") # used for LSTM
+tokenizer = joblib.load("tokenizer.joblib") # used for LSTM
 
 # Decode label function
 # {'business': 0, 'entertainment': 1, 'health': 2, 'politics': 3, 'sport': 4}
@@ -78,24 +78,26 @@ def crawURL(url):
 def process_api(text):
     # Vectorize the text data
     processed_text = vectorizer.transform([text])
-    # sequence = tokenizer.texts_to_sequences([text])
-    # padded_sequence = pad_sequences(sequence, maxlen=1000, padding='post')
+    sequence = tokenizer.texts_to_sequences([text])
+    padded_sequence = pad_sequences(sequence, maxlen=1000, padding='post')
     
     # Get the predicted result from models
     Logistic_Predicted = logistic_model.predict(processed_text).tolist() # Logistic Model
     SVM_Predicted = SVM_model.predict(processed_text).tolist() # SVC Model
-    # Seq_Predicted = Seq_model.predict(padded_sequence)
+    Seq_Predicted = Seq_model.predict(padded_sequence)
     # predicted_label_index = np.argmax(Seq_Predicted)
 
     # ----------- Debug Logs -----------
     logistic_debug = decodedLabel(int(Logistic_Predicted[0]))
     svc_debug = decodedLabel(int(SVM_Predicted[0]))
+    predicted_label_index = np.argmax(Seq_Predicted)
     print('Logistic', int(Logistic_Predicted[0]), logistic_debug)
     print('SVM', int(SVM_Predicted[0]), svc_debug)
     
     return {
             'Logistic_Predicted':decodedLabel(int(Logistic_Predicted[0])),
             'SVM_Predicted': decodedLabel(int(SVM_Predicted[0])),
+            'LSTM': decodedLabel(int(predicted_label_index)),
             'Article_Content': text
         }