使用docker训练时报错Check failed: cudaSuccess == err (0 vs. 8)
Created by: zongweiLi
F0123 11:55:36.803860 19 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: *
@ 0x7fbf82b2a04d google::LogMessage::Fail()
@ 0x7fbf82b2c398 google::LogMessage::SendToLog()
@ 0x7fbf82b29b5b google::LogMessage::Flush()
@ 0x7fbf82b2d26e google::LogMessageFatal::~LogMessageFatal()
@ 0x7fbf8296d9df hl_gpu_apply_unary_op<>()
@ 0x7fbf8296de1f paddle::BaseMatrixT<>::applyUnary<>()
@ 0x7fbf8296e0e8 paddle::BaseMatrixT<>::zero()
@ 0x7fbf828de8b4 paddle::GpuMatrix::zeroMem()
@ 0x7fbf8278e421 paddle::BatchNormBaseLayer::init()
@ 0x7fbf8272fa62 paddle::CudnnBatchNormLayer::init()
@ 0x7fbf827b5d78 paddle::NeuralNetwork::init()
@ 0x7fbf827c938a paddle::MultiGradientMachine::MultiGradientMachine()
@ 0x7fbf827ce1de paddle::GradientMachine::create()
@ 0x7fbf82afa8bb GradientMachine::createFromPaddleModelPtr()
@ 0x7fbf82afb4ce GradientMachine::createByConfigProtoStr()
@ 0x7fbf825e23aa _wrap_GradientMachine_createByConfigProtoStr
@ 0x4cb45e PyEval_EvalFrameEx
@ 0x4c2765 PyEval_EvalCodeEx
@ 0x4ca8d1 PyEval_EvalFrameEx
@ 0x4c2765 PyEval_EvalCodeEx
@ 0x4ca099 PyEval_EvalFrameEx
@ 0x4c2765 PyEval_EvalCodeEx
@ 0x4de8b8 (unknown)
@ 0x4b0cb3 PyObject_Call
@ 0x4f492e (unknown)
@ 0x4b0cb3 PyObject_Call
@ 0x4f46a7 (unknown)
@ 0x4b670c (unknown)
@ 0x4b0cb3 PyObject_Call
@ 0x4c9faf PyEval_EvalFrameEx
@ 0x4c2765 PyEval_EvalCodeEx
@ 0x4ca099 PyEval_EvalFrameEx
Aborted (core dumped)
Failed in training!
