use cpu to run model

chatglm2_6b/modelClient.py

@@ -22,8 +22,7 @@ class ChatGLM2(object):
         if model_path:
             self.model_path = model_path
         self.tokenizer = AutoTokenizer.from_pretrained(self.model_path, trust_remote_code=True)
-        model = AutoModel.from_pretrained(self.model_path, trust_remote_code=True).half().cuda()
-        self.model = model.eval()
+        self.model = AutoModel.from_pretrained(self.model_path, trust_remote_code=True).float()
 
     def generate(
         self,