Created by: waywaywayw

1）PaddleHub和PaddlePaddle版本： paddlehub 1.5.4 paddlepaddle-gpu 1.7.0.post107

2）系统环境：请您描述系统类型，例如Linux/Windows/MacOS/，python版本 拉取docker镜像 FROM hub.baidubce.com/paddlepaddle/paddle:1.7.0-gpu-cuda10.0-cudnn7 使用镜像里的python3.6

3）复现信息：如为报错，请给出复现环境、复现步骤 已设置CUDA_VISIBLE_DEVICES=0 已跑通demo，使用如下语句时，速度明显比cpu变快，gpu调用成功。module.lexical_analysis(data=inputs, use_gpu=True, batch_size=64)

尝试使用 hub serving 提供gpu服务。 开启服务：hub serving start --config config.json

config.json里设置"use_gpu": false 时，可以正常运行 设置"use_gpu": true 时报错如下：

--------------------------------------------
C++ Call Stacks (More useful to developers):
--------------------------------------------
0   std::string paddle::platform::GetTraceBackString<char const*>(char const*&&, char const*, int)
1   paddle::platform::EnforceNotMet::EnforceNotMet(std::__exception_ptr::exception_ptr, char const*, int)
2   paddle::platform::SetDeviceId(int)
3   paddle::memory::allocation::CUDAAllocator::AllocateImpl(unsigned long)
4   paddle::memory::allocation::AlignedAllocator::AllocateImpl(unsigned long)
5   paddle::memory::allocation::AutoGrowthBestFitAllocator::AllocateImpl(unsigned long)
6   paddle::memory::allocation::RetryAllocator::AllocateImpl(unsigned long)
7   paddle::memory::allocation::AllocatorFacade::Alloc(paddle::platform::Place const&, unsigned long)
8   paddle::memory::allocation::AllocatorFacade::AllocShared(paddle::platform::Place const&, unsigned long)
9   paddle::memory::AllocShared(paddle::platform::Place const&, unsigned long)
10  paddle::framework::Tensor::mutable_data(paddle::platform::Place const&, paddle::framework::proto::VarType_Type, unsigned long)

----------------------
Error Message Summary:
----------------------
Error: cudaSetDevice failed in paddle::platform::SetDeviced, error code : 3, Please see detail in https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html#group__CUDART__TYPES_1g3f51e3575c2178246db0a94a430e0038: initialization error at (/paddle/paddle/fluid/platform/gpu_info.cc:240)