Update app.py

app.py

@@ -11,7 +11,7 @@ model = None
 tokenizer = None
 generator = None
 
-def load_model(model_name, eight_bit=0, device_map="auto"):
+def load_model(model_name = "zl111/ChatDoctor", eight_bit=0, device_map="auto"):
     global model, tokenizer, generator
 
     print("Loading "+model_name+"...")
@@ -24,7 +24,7 @@ def load_model(model_name, eight_bit=0, device_map="auto"):
     print('gpu_count', gpu_count)
 
     tokenizer = transformers.LlamaTokenizer.from_pretrained(model_name)
-    model = transformers.LLaMAForCausalLM.from_pretrained(
+    model = transformers.LlamaForCausalLM.from_pretrained(
         model_name,
         #device_map=device_map,
         #device_map="auto",