search

PaddlePaddle / PaddleOCR

Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices)
https://paddlepaddle.github.io/PaddleOCR/
Apache License 2.0
42.98k stars 7.72k forks source link

Segmentation fault when running inference model #8612

Closed mazatov closed 1 year ago

mazatov commented 1 year ago

I'm running the code on AWS ubuntu machine. I am using paddlepaddle-gpu=2.4.1 . My cuda version is: 

(paddle_env) ubuntu@ip-10-0-11-36:~/MVTools$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2018 NVIDIA Corporation
Built on Sat_Aug_25_21:08:01_CDT_2018
Cuda compilation tools, release 10.0, V10.0.130

I was able to train my model and create an inference model that I have tested on Windows and it works well. I'm trying to use this inference model on the same Ubuntu machine and am getting this error.

I've read in other issues that it might be something to do with cuda version. I need to keep cuda 10.0 for other applications. If it indeed needs a different cuda version, can I add it alongside the 10.0? 

--------------------------------------
C++ Traceback (most recent call last):
--------------------------------------
No stack trace in paddle, may be caused by external reasons.

----------------------
Error Message Summary:
----------------------
FatalError: `Segmentation fault` is detected by the operating system.
  [TimeInfo: *** Aborted at 1670953974 (unix time) try "date -d @1670953974" if you are using GNU date ***]
  [SignalInfo: *** SIGSEGV (@0x0) received by PID 5741 (TID 0x7f28e892d700) from PID 0 ***]

Segmentation fault (core dumped)
github-actions[bot] commented 1 year ago

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 7 days if no further activity occurs. Thank you for your contributions.