Serving cuda error

用paddle inference 不会报错，用pipline serving会报错

CRITICAL 2022-02-09 10:46:11,025 [operator.py:1315] [ner|1] failed to init op: (External) CUDA error(3), initialization error. [Hint: 'cudaErrorInitializationError'. The API call failed because the CUDA driver and runtime could not be initialized. ] (at /paddle/paddle/fluid/platform/gpu_info.cc:355) Traceback (most recent call last): File "/root/anaconda3/envs/paddle_env/lib/python3.7/site-packages/paddle_serving_server/pipeline/operator.py", line 1308, in _run concurrency_idx) File "/root/anaconda3/envs/paddle_env/lib/python3.7/site-packages/paddle_serving_server/pipeline/local_service_handler.py", line 228, in get_client mkldnn_bf16_op_list=self._mkldnn_bf16_op_list) File "/root/anaconda3/envs/paddle_env/lib/python3.7/site-packages/paddle_serving_app/local_predict.py", line 235, in load_model_config self.predictor = paddle_infer.create_predictor(config) OSError: (External) CUDA error(3), initialization error. [Hint: 'cudaErrorInitializationError'. The API call failed because the CUDA driver and runtime could not be initialized. ] (at /paddle/paddle/fluid/platform/gpu_info.cc:355)