Created by: yeyupiaoling

• 我使用的是百度深度学习的GPU服务器，更新到最新的PaddlePaddle的版本了
• 我查看了CUDA的版本，为CUDA Version 8.0.61，CDNN的版本是5.1.10
• 我还要安装什么吗？
• 报错如下：

I0202 13:29:53.189960 28490 Util.cpp:166] commandline:  --use_gpu=True --trainer_count=1 
F0202 13:29:53.192677 28490 hl_cuda_device.cc:453] Check failed: cudaSuccess == cudaStat (0 vs. 38) Cuda Error: no CUDA-capable device is detected
*** Check failure stack trace: ***
    @     0x7f9db2b88bcd  google::LogMessage::Fail()
    @     0x7f9db2b8c67c  google::LogMessage::SendToLog()
    @     0x7f9db2b886f3  google::LogMessage::Flush()
    @     0x7f9db2b8db8e  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f9db2b4455f  hl_specify_devices_start()
    @     0x7f9db2b4485d  hl_start()
    @     0x7f9db2ace55a  paddle::initMain()
    @     0x7f9db2b705d1  initPaddle()
    @     0x7f9db2717c29  _wrap_initPaddle
    @           0x4cb755  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4ca7df  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4de69e  (unknown)
    @           0x4b0c93  PyObject_Call
    @           0x4c6ef6  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4ca7df  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4de858  (unknown)
    @           0x4b0c93  PyObject_Call
    @           0x4c6ef6  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4de858  (unknown)
    @           0x4b0c93  PyObject_Call
    @           0x4c6ef6  PyEval_EvalFrameEx
    @           0x4c9d7f  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
    @           0x4de858  (unknown)
    @           0x4b0c93  PyObject_Call
    @           0x4c6ef6  PyEval_EvalFrameEx
    @           0x4c2705  PyEval_EvalCodeEx
Aborted (core dumped)