推理ocr时候出现错误

[wangning7149]

运行，命令 python tools/infer/predict_system.py --image_dir="./doc/imgs/4.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/"时，出现以下错误：

Traceback (most recent call last):
File "tools/infer/predict_system.py", line 115, in
dt_boxes, rec_res = text_sys(img)
File "tools/infer/predict_system.py", line 78, in call
rec_res, elapse = self.text_recognizer(img_crop_list)
File "/home/wn/projects/PaddleOCR-develop/tools/infer/predict_rec.py", line 82, in call
self.predictor.zero_copy_run()
paddle.fluid.core_avx.EnforceNotMet:

---

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::operators::math::FCFunctor<paddle::platform::CUDADeviceContext, float>::operator()(paddle::platform::CUDADeviceContext const&, int, int, int, float const*, float const*, float*, float const*, bool, bool)
3 paddle::operators::FCOpKernel<paddle::platform::CUDADeviceContext, float>::Compute(paddle::framework::ExecutionContext const&) const
4 std::_Function_handler<void (paddle::framework::ExecutionContext const&), paddle::framework::OpKernelRegistrarFunctor<paddle::platform::CUDAPlace, false, 0ul, paddle::operators::FCOpKernel<paddle::platform::CUDADeviceContext, float>, paddle::operators::FCOpKernel<paddle::platform::CUDADeviceContext, double> >::operator()(char const*, char const*, int) const::{lambda(paddle::framework::ExecutionContext const&)#1}>::_M_invoke(std::_Any_data const&, paddle::framework::ExecutionContext const&)
5 paddle::framework::OperatorWithKernel::RunImpl(paddle::framework::Scope const&, paddle::platform::Place const&, paddle::framework::RuntimeContext*) const
6 paddle::framework::OperatorWithKernel::RunImpl(paddle::framework::Scope const&, paddle::platform::Place const&) const
7 paddle::framework::OperatorBase::Run(paddle::framework::Scope const&, paddle::platform::Place const&)
8 paddle::framework::NaiveExecutor::Run()
9 paddle::AnalysisPredictor::ZeroCopyRun()

---

Error Message Summary:

Error: An error occurred here. There is no accurate error hint for this error yet. We are continuously in the process of increasing hint for this kind of error check. It would be helpful if you could inform us of how this conversion went by opening a github issue. And we will resolve it with high priority.

• New issue link: https://github.com/PaddlePaddle/Paddle/issues/new
• Recommended issue content: all error stack information
  [Hint: CUBLAS_STATUS_EXECUTION_FAILED] at (/paddle/paddle/fluid/operators/math/blas_impl.cu.h:34)

请问是什么原因呢？

[LDOUBLEV]

像是cuda的问题，可以试下能否在CPU上运行:

python tools/infer/predict_system.py --image_dir="./doc/imgs/4.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/" --use_gpu=False

你安装的是哪个版本的paddle，可否尝试Paddle 1.7.0， Paddle 1.7.2；

另外确认下你的cuda版本和机器的显卡以及cudnn版本是否匹配吧！

如果可以的话，建议直接使用我们提供的docker镜像

[wangning7149]

像是cuda的问题，可以试下能否在CPU上运行:
python tools/infer/predict_system.py --image_dir="./doc/imgs/4.jpg" --det_model_dir="./inference/det_db/" --rec_model_dir="./inference/rec_crnn/" --use_gpu=False

你安装的是哪个版本的paddle，可否尝试Paddle 1.7.0， Paddle 1.7.2；
另外确认下你的cuda版本和机器的显卡以及cudnn版本是否匹配吧！
如果可以的话，建议直接使用我们提供的docker镜像

cpu 运行是可以的。
paddle版本：paddlepaddle-gpu 1.7.2.post107
CUDA版本：CUDA Version 10.1.243
cudnn版本：
#define CUDNN_MAJOR 7
#define CUDNN_MINOR 6
#define CUDNN_PATCHLEVEL 4

#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

显卡：RTX-2080

[henuliulei]

请问解决了吗