Update app.py

app.py

@@ -1,6 +1,4 @@
 from transformers import pipeline
-import gradio as gr # Import Gradio for the interface
-
 
 # Load a text-generation model
 chatbot = pipeline("text-generation", model="microsoft/DialoGPT-medium")
@@ -10,14 +8,15 @@ classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnl
 
 # Customize the bot's knowledge base with predefined responses
 faq_responses = {
-    "study tips": "Here are some study tips: 1) Break your study sessions into 25-minute chunks (Pomodoro Technique). 2) Test yourself frequently. 3) Stay organized using planners or apps like Notion or Todoist.",
+    "study tips": "Here are some study tips: 1) Break your study sessions into 25-minute chunks (Pomodoro Technique). 2) Test yourself frequently. 3) Stay organized using planners or apps like Notion or Todoist."
+}
 
 # Define the chatbot's response function
 def faq_chatbot(user_input):
     # Classify the user input by passing the FAQ keywords as labels
     classified_user_input = classifier(user_input, candidate_labels=list(faq_responses.keys()))
 
-    # Get the highest confidence score label, ie. the most likely of the FAQ
+    # Get the highest confidence score label, i.e., the most likely FAQ
     predicted_label = classified_user_input["labels"][0]
     confidence_score = classified_user_input["scores"][0]
 
@@ -27,12 +26,9 @@ def faq_chatbot(user_input):
     # If the classification confidence is high, return the corresponding FAQ response
     if confidence_score > threshold:
         return faq_responses[predicted_label]
-    
-    
-    # Check if the user's input matches any FAQ keywords
-    # for key, response in faq_responses.items():
-    #     if key in user_input.lower():
-    #         return response
-
- # If no FAQ match, use the AI model to generate a response
+
+    # If no FAQ match, use the AI model to generate a response
     conversation = chatbot(user_input, max_length=50, num_return_sequences=1)
+    return conversation[0]["generated_text"]
+
+