Update app.py

app.py

@@ -29,10 +29,10 @@ def askme(symptoms, question):
     prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
     inputs = tokenizer(prompt, return_tensors="pt").to(device)  # Ensure inputs are on CUDA device
     outputs = model.generate(**inputs, max_new_tokens=200, use_cache=True)
-    response_text = tokenizer.batch_decode(outputs)[0].strip() #skip_special_tokens=True
+    response_text = tokenizer.batch_decode(outputs, skip_special_tokens=False)[0].strip() #skip_special_tokens=True
     # Remove system messages and content
     # Extract only the assistant's response
-    assistant_response = response_text.split("<|im_start|>assistant")[1].strip().replace('<|im_end', '')
+    assistant_response = response_text.split("<|im_start|>assistant")[1].strip().replace('<|im_end|>', '')
    #  Extract only the assistant's response
     return assistant_response