Update app.py

app.py

@@ -7,6 +7,7 @@ from transformers import T5Tokenizer, T5ForConditionalGeneration
 # Load pre-trained model and tokenizer
 tokenizer = T5Tokenizer.from_pretrained("ahmed792002/Finetuning_T5_HealthCare_Chatbot")
 model = T5ForConditionalGeneration.from_pretrained("ahmed792002/Finetuning_T5_HealthCare_Chatbot")
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 # Function to clean input text
 def clean_text(text):
@@ -17,14 +18,16 @@ def clean_text(text):
     return text
 
 # Chatbot function
-def chatbot(query):
+def chatbot(query, history, system_message):
     query = clean_text(query)
     input_ids = tokenizer(query, return_tensors="pt", max_length=256, truncation=True)
     inputs = {key: value.to(device) for key, value in input_ids.items()}
     outputs = model.generate(
         input_ids["input_ids"],
-        max_length=1024,
+        max_length=1024,  # Adjust this as needed for your use case
         num_beams=5,
+        temperature=0.7,  # Adjust this as needed
+        top_p=0.95,  # Adjust this as needed
         early_stopping=True
     )
     return tokenizer.decode(outputs[0], skip_special_tokens=True)
@@ -40,4 +43,4 @@ demo = gr.ChatInterface(
 )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(share=True)  # Set `share=True` to create a public link