Update app.py

app.py

@@ -1,25 +1,46 @@
 import gradio as gr
-from transformers import pipeline
+from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 
-# Load a chatbot model from Hugging Face
-chatbot_pipeline = pipeline("conversational", model="microsoft/DialoGPT-medium")
+# Load the advanced conversational model (e.g., GPT-NeoX-20B or GPT-J-6B)
+model_name = "EleutherAI/gpt-j-6B"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+chatbot_pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer, max_length=200)
 
-# Function to generate a response
-def respond(user_input, history=[]):
-    conversation = chatbot_pipeline.create_conversation(user_input)
-    chatbot_pipeline(conversation)
-    response = conversation.generated_responses[-1]
+# Define FAQs for instant responses
+faqs = {
+    "How do I enroll in a course?": "To enroll, go to our website, select the course you're interested in, and click 'Enroll Now'. Follow the on-screen instructions.",
+    "What is the refund policy?": "We offer a full refund within the first 14 days if you are not satisfied with the course.",
+    "How can I access my course materials?": "Once you enroll, you can access course materials in the 'My Courses' section of your account.",
+    "Are there any live sessions available?": "Yes, many courses include live sessions. You can view the schedule in the course overview.",
+    "How can I get a certificate?": "Complete all modules and pass the final assessment with a minimum score of 70% to receive your certificate.",
+    # Add more FAQs as needed
+}
+
+# Function to generate a response with FAQs and model for other queries
+def generate_response(user_input, history=[]):
+    # Check if input matches any FAQ
+    if user_input in faqs:
+        response = faqs[user_input]
+    else:
+        # Generate response using the model for non-FAQ queries
+        prompt = f"Student: {user_input}\nAssistant:"
+        model_response = chatbot_pipeline(prompt)[0]['generated_text']
+        response = model_response.split("Assistant:")[-1].strip()
+    
     history.append((user_input, response))
     return "", history
 
-# Interface for the chatbot using Gradio
+# Real-time chatbot UI with Gradio
 iface = gr.Interface(
-    fn=respond,
+    fn=generate_response,
     inputs=["text", "state"],
     outputs=["text", "state"],
     live=True,
-    title="Hugging Face Chatbot",
-    description="A simple chatbot using Hugging Face's DialoGPT model. Type your message and press Enter to get a response!",
+    title="Edutech Student Chatbot",
+    description="Ask me anything about courses, enrollment, certification, and more!",
+    theme="compact",
+    css=".gradio-container { background-color: #f5f5f5; font-family: Arial, sans-serif; }"
 )
 
 # Launch the chatbot interface