Created by: zmofei

我在使用GPU训练SSD模型的时候（models/ssd/trainer.py）, 遇到如下错误：

F0921 14:24:18.580162  6954 hl_gpu_matrix_kernel.cuh:181] Check failed: cudaSuccess == err (0 vs. 8) [hl_gpu_apply_unary_op failed] CUDA error: invalid device function
*** Check failure stack trace: ***
    @     0x7fcdf3faf5ed  google::LogMessage::Fail()
    @     0x7fcdf3fb309c  google::LogMessage::SendToLog()
    @     0x7fcdf3faf0e3  google::LogMessage::Flush()
    @     0x7fcdf3fb45ae  google::LogMessageFatal::~LogMessageFatal()
    @     0x7fcdf3e39ec4  hl_gpu_apply_unary_op<>()
    @     0x7fcdf3e3a205  paddle::BaseMatrixT<>::applyUnary<>()
    @     0x7fcdf3e3a433  paddle::BaseMatrixT<>::zero()
    @     0x7fcdf3cd58d1  paddle::Parameter::enableType()
    @     0x7fcdf3cd11cc  paddle::parameterInitNN()
    @     0x7fcdf3cd391a  paddle::NeuralNetwork::init()
    @     0x7fcdf3cfc491  paddle::GradientMachine::create()
    @     0x7fcdf3f8c3b3  GradientMachine::createFromPaddleModelPtr()
    @     0x7fcdf3f8c58f  GradientMachine::createByConfigProtoStr()
    @     0x7fcdf3b9b4cd  _wrap_GradientMachine_createByConfigProtoStr
    @           0x4b4cb9  PyEval_EvalFrameEx
    @           0x4b6b28  PyEval_EvalCodeEx
    @           0x4b5d10  PyEval_EvalFrameEx
    @           0x4b6b28  PyEval_EvalCodeEx
    @           0x4b5d10  PyEval_EvalFrameEx
    @           0x4b6b28  PyEval_EvalCodeEx
    @           0x52940f  function_call
    @           0x422cba  PyObject_Call
    @           0x4271ad  instancemethod_call
    @           0x422cba  PyObject_Call
    @           0x48121f  slot_tp_init
    @           0x47eb1a  type_call
    @           0x422cba  PyObject_Call
    @           0x4b31dd  PyEval_EvalFrameEx
    @           0x4b6b28  PyEval_EvalCodeEx
    @           0x4b5d10  PyEval_EvalFrameEx
    @           0x4b6b28  PyEval_EvalCodeEx
    @           0x4b6c52  PyEval_EvalCode
Aborted

另外，该模型修改后在CPU下是可以运行的。 请问如何解决这个问题？我是使用pip源安装的