Update app.py

app.py

@@ -1,5 +1,5 @@
 from fastai.text.all import *
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import BlenderbotForConditionalGeneration, BlenderbotTokenizer
 import torch
 import gradio as gr
 
@@ -31,20 +31,20 @@ def classify_psychiatric_text(txt):
     probabilities = torch.softmax(logits, dim=1).squeeze().tolist()
     return dict(zip(psychiatric_labels, probabilities))
 
-# Load pre-trained conversational model for Lifestyle and Nutrition Chatbot
-lifestyle_model_name = "microsoft/DialoGPT-medium"  # Replace with a fine-tuned model if available
-lifestyle_tokenizer = AutoTokenizer.from_pretrained(lifestyle_model_name)
-lifestyle_model = AutoModelForCausalLM.from_pretrained(lifestyle_model_name)
+# Load BlenderBot for Lifestyle and Nutrition Chatbot
+blender_model_name = "facebook/blenderbot-3B"  # Pre-trained BlenderBot 3B model
+blender_tokenizer = BlenderbotTokenizer.from_pretrained(blender_model_name)
+blender_model = BlenderbotForConditionalGeneration.from_pretrained(blender_model_name)
 
 # Chat function for Lifestyle and Nutrition
 chat_history = []
 
 def chatbot_response(user_input):
     global chat_history
-    new_input_ids = lifestyle_tokenizer.encode(user_input + lifestyle_tokenizer.eos_token, return_tensors='pt')
+    new_input_ids = blender_tokenizer.encode(user_input + blender_tokenizer.eos_token, return_tensors='pt')
     bot_input_ids = torch.cat([chat_history, new_input_ids], dim=-1) if chat_history else new_input_ids
-    chat_history = lifestyle_model.generate(bot_input_ids, max_length=1000, pad_token_id=lifestyle_tokenizer.eos_token_id)
-    response = lifestyle_tokenizer.decode(chat_history[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)
+    chat_history = blender_model.generate(bot_input_ids, max_length=1000, pad_token_id=blender_tokenizer.eos_token_id)
+    response = blender_tokenizer.decode(chat_history[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)
     return response
 
 def clear_chat():