Erroneous arithmetic operation is detected by operating system

[Vaishal16]

Using PP-OCR english recognition model for training purpose.
Model name : en_number_mobile_v2.0_rec

I am getting the following error while training on the above pretrained model:

C++ Traceback (most recent call last):
0 paddle::imperative::Tracer::TraceOp(std::string const&, paddle::imperative::NameVarBaseMap const&, paddle::imperative::NameVarBaseMap const&, paddle::framework::AttributeMap, std::map<std::string, std::string, std::less<std::string >, std::allocator<std::pair<std::string const, std::string > > > const&)
1 paddle::imperative::Tracer::TraceOp(std::string const&, paddle::imperative::NameVarBaseMap const&, paddle::imperative::NameVarBaseMap const&, paddle::framework::AttributeMap, paddle::platform::Place const&, bool, std::map<std::string, std::string, std::less<std::string >, std::allocator<std::pair<std::string const, std::string > > > const&)
2 paddle::imperative::PreparedOp::Run(paddle::imperative::NameVarBaseMap const&, paddle::imperative::NameVarBaseMap const&, paddle::framework::AttributeMap const&)
3 std::_Function_handler<void (paddle::framework::ExecutionContext const&), paddle::framework::OpKernelRegistrarFunctor<paddle::platform::CUDAPlace, false, 0ul, paddle::operators::WarpCTCKernel<paddle::platform::CUDADeviceContext, float>, paddle::operators::WarpCTCKernel<paddle::platform::CUDADeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1 (closed)}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&)
4 paddle::operators::WarpCTCKernel<paddle::platform::CUDADeviceContext, float>::Compute(paddle::framework::ExecutionContext const&) const
5 paddle::operators::math::UnpaddingLoDTensorFunctor<paddle::platform::CUDADeviceContext, int>::operator()(paddle::platform::CUDADeviceContext const&, paddle::framework::LoDTensor const&, paddle::framework::LoDTensor*, int, int, bool, paddle::operators::math::PadLayout)
6 paddle::framework::SignalHandle(char const*, int)
7 paddle::platform::GetCurrentTraceBackStringabi:cxx11

Error Message Summary:
FatalError: Erroneous arithmetic operation is detected by the operating system.
[TimeInfo: *** Aborted at 1634193209 (unix time) try "date -d @1634193209" if you are using GNU date ***]
[SignalInfo: *** SIGFPE (@0x7fc3b6a3a120) received by PID 32157 (TID 0x7fc3ff43d100) from PID 18446744072478761248 ***]

Please help me out with the issue.

[tink2123]

Could you provide the following details:

System information

• PaddlePaddle version （eg.1.1）or CommitID
• CPU: including CPUMKL/OpenBlas/MKLDNN version
• GPU: including CUDA/CUDNN version
• OS Platform (eg.Mac OS 10.14)

To Reproduce
Describe your current behavior
Code to reproduce the issue

According to the current error message, it may be that the CUDA environment is installed incorrectly. It is recommended to use it in docker.

You can refer to installation

[Vaishal16]

CommitID : 5071292 (branch : release/2.3)
paddlepaddle-gpu==2.0.0
CUDNN version 10.0
OS : Ubuntu 18.04

To reprodce:

Code used if from branch : release/2.3.
Model and config files used are of English Recognition Model from here

Meanwhile trying out docker installation and setting up CUDA environment again.

[Vaishal16]

Worked with proper CUDA version installation in Docker.

Closing the issue now.