Update app.py

app.py

@@ -5,18 +5,20 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 
 # Load MedScholar model and tokenizer
 model_name = "yasserrmd/MedScholar-1.5B"
-tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    device_map="auto",
-    torch_dtype=torch.float16,
-    trust_remote_code=True
-)
-model.eval()
+
+
 
 # Chat function (streaming style)
 @spaces.GPU
 def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, top_p):
+    tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
+    model = AutoModelForCausalLM.from_pretrained(
+        model_name,
+        device_map="auto",
+        torch_dtype=torch.float16,
+        trust_remote_code=True
+    )
+    model.eval()
     # Prepare the full conversation
     conversation = [{"role": "system", "content": system_message}]
     for user_msg, bot_reply in history: