폐쇄망 환경에서 cuda 및 torch를 설치 후, Transformers의 Trainer class를 활용해 학습을 진행할 때
발생하는 메세지로, C++ 관련 백엔드 부분에 문제가 생기는 것으로 보이는데, 아직 정확한 해결책은 찾지 못한 상황
참고해볼만한 링크들
https://discuss.pytorch.org/t/malloc-and-corrupted-size-errors/37080
Malloc, and corrupted size errors
Hi. I’m using pytorch in multi gpu setting, and also with some multiprocessing for loading and preprocessing and quite often encountering these errors, which I can’t manage to solve. malloc.c:4023: _int_malloc: Assertion (unsigned long) (size) >= (unsi
discuss.pytorch.org
https://github.com/pytorch/pytorch/issues/2507
Segfault (free() on invalid pointer) · Issue #2507 · pytorch/pytorch
On Ubuntu 16.04, Python 3.6.2 and PyTorch 0.2.0, it's possible to force a free() on invalid pointer segfault by first doing import zmq then import torch. (Originally, I had this problem when tr...
github.com
https://github.com/pytorch/pytorch/issues/10567
problem: corrupted size vs. prev_size when using python3.6 · Issue #10567 · pytorch/pytorch
Hi, I install tensorflow-1.9; then install pytorch=0.4 at python2 and python3 respectively. I use python2 to run my code, no error is reported; but when using python3 to run it, an error is report ...
github.com
https://github.com/pytorch/pytorch/issues/26467
Pytorch 1.2 on Red Hat Enterprise Linux 6.4 gives 'GLIBC_2.14' not found · Issue #26467 · pytorch/pytorch
🐛 Bug Description: On a server running Red Hat Enterprise Linux (RHEL) Workstaion release 6.4 (Santiago), I used conda 4.5.12 to install pytorch using the following command: conda install pytorch t...
github.com
'23년 이전 글 > Error Today' 카테고리의 다른 글
| TPU 에러 모음 (0) | 2022.11.26 |
|---|---|
| 서비스 서버에 대한 주의 (0) | 2022.08.04 |
| Pytorch train 과정에서 발생한 OOM(out of memory) (0) | 2022.08.03 |