Merge branch 'v0.6.2-dev' into 'v0.6.2-dev'

Update offline_streaming_inference_chat_demo.py

See merge request dcutoolkit/deeplearing/vllm!49

examples/offline_streaming_inference_chat_demo.py

 
 '''
-python offline_streaming_inference_chat_demo.py --model /models/llama2/Llama-2-7b-chat-hf  --template template_llama_chat.jinja --dtype float16 --enforce-eager -tp 1 
+python offline_streaming_inference_chat_demo.py --model /models/llama2/Llama-2-7b-chat-hf  --dtype float16 --enforce-eager -tp 1 
 '''
 from vllm.sampling_params import SamplingParams
 from vllm.engine.async_llm_engine import AsyncEngineArgs, AsyncLLMEngine
@@ -36,7 +36,6 @@ class FlexibleArgumentParser(argparse.ArgumentParser):
         return super().parse_args(processed_args, namespace)
    
 parser = FlexibleArgumentParser()
-parser.add_argument('--template', type=str, help="Path to template")
 parser = AsyncEngineArgs.add_cli_args(parser)
 args = parser.parse_args()
 
@@ -47,13 +46,13 @@ args = parser.parse_args()
 # ]
 
 tokenizer =  AutoTokenizer.from_pretrained(args.model)
-try:
-     f = open(args.template,'r')
-     tokenizer.chat_template = f.read()
-except Exception as e:
-     print('except:',e)
-finally:
-     f.close()
+# try:
+#      f = open(args.template,'r')
+#      tokenizer.chat_template = f.read()
+# except Exception as e:
+#      print('except:',e)
+# finally:
+#      f.close()
 
 
 
@@ -110,4 +109,3 @@ while True:
      history.append({"role": "assistant", "content": response})
 print()
 
-