Update app.py

app.py

@@ -2,65 +2,52 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 import gradio as gr
 import torch
 import json
-
+import re
 
 title = "????AI ChatBot"
 description = "A State-of-the-Art Large-scale Pretrained Response generation model (DialoGPT)"
-examples = [["What courses are available in Engineering?"]]
-
+examples = [["What courses under Engineering?"]]
 
 tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-large")
 model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-large")
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-
-model.to(device)
 
 # Load courses data from JSON file
-with open("uts_courses.json", "r") as f:
-    courses_data = json.load(f)
+def load_courses(file_path):
+    with open(file_path, 'r') as file:
+        courses_data = json.load(file)
+    return courses_data
 
+courses_data = load_courses('uts_courses.json')
 
 def predict(input, history=[]):
-    # Check if the input question is about courses
-    if "courses" in input.lower():
-        # Check if the input question contains a specific field (e.g., Engineering, Information Technology, etc.)
-        for field in courses_data["courses"]:
-            if field.lower() in input.lower():
-                # Get the list of courses for the specified field
-                courses_list = courses_data["courses"][field]
-                # Format the response
-                response = f"The available courses in {field} are: {', '.join(courses_list)}."
-                return response, history
-
-    # If the input question is not about courses, use the dialogue model to generate a response
-    # tokenize the new input sentence
-    new_user_input_ids = tokenizer.encode(
-        input + tokenizer.eos_token, return_tensors="pt"
-    ).to(device)
-
-    # append the new user input tokens to the chat history
-    bot_input_ids = torch.cat([torch.tensor(history).to(device), new_user_input_ids], dim=-1)
-
-    # generate a response
-    history = model.generate(
-        bot_input_ids, max_length=4000, pad_token_id=tokenizer.eos_token_id
-    ).tolist()
-
-    # convert the tokens to text, and then split the responses into lines
-    response = tokenizer.decode(history[0]).split("")
-    response = [
-        (response[i], response[i + 1]) for i in range(0, len(response) - 1, 2)
-    ]  # convert to tuples of list
-    return response, history
-
-
-def main():
-    pass
-
-
-if __name__ == "__main__":
-    main()
-
+    # Check if the input is related to courses
+    if re.match(r'what courses under (\w+)', input.strip().lower()):
+        # Extract the field of interest (e.g., Engineering, Information Technology, etc.)
+        match = re.match(r'what courses under (\w+)', input.strip().lower())
+        field = match.group(1).capitalize()  # Capitalize the field name
+        
+        if field in courses_data['courses']:
+            # Get the list of courses for the specified field
+            courses = courses_data['courses'][field]
+            response = f"The available courses in {field} are: {', '.join(courses)}"
+        else:
+            response = "I'm sorry, I couldn't find any courses related to that field."
+
+        return response, history
+    else:
+        # Tokenize the new input sentence
+        new_user_input_ids = tokenizer.encode(input + tokenizer.eos_token, return_tensors="pt")
+
+        # Append the new user input tokens to the chat history
+        bot_input_ids = torch.cat([torch.LongTensor(history), new_user_input_ids], dim=-1)
+
+        # Generate a response
+        history = model.generate(bot_input_ids, max_length=4000, pad_token_id=tokenizer.eos_token_id).tolist()
+
+        # Convert the tokens to text
+        response = tokenizer.decode(history[0])
+
+        return response, history
 
 gr.Interface(
     fn=predict,
@@ -70,5 +57,4 @@ gr.Interface(
     inputs=["text", "state"],
     outputs=["chatbot", "state"],
     theme="finlaymacklon/boxy_violet",
-    share=True,
 ).launch()