PaddleNLP
PaddleNLP copied to clipboard
[Bug]: applications-text_summarization-pegasus使用run_train.py出现问题,环境版本与aistudio一致,无法运行
软件环境
- paddlepaddle:
- paddlepaddle-gpu: 2.3.2
- paddlenlp: 2.4.2
重复问题
- [X] I have searched the existing issues
错误描述
_init_parallel_ctx
__parallel_ctx__clz__.init()
OSError: (External) NCCL error(1), unhandled cuda error.
[Hint: 'ncclUnhandledCudaError'. A call to a CUDA function failed.] (at /paddle/paddle/fluid/platform/collective_helper.cc:100)
LAUNCH INFO 2022-11-10 15:08:16,662 Exit code 1
INFO 2022-11-10 15:08:16,662 controller.py:124] Exit code 1
稳定复现步骤 & 代码
sh run_train.sh
你好,这个本地环境的问题,可以参考这个更新下:https://github.com/PaddlePaddle/PaddleDetection/issues/4139
This issue is stale because it has been open for 60 days with no activity. 当前issue 60天内无活动,被标记为stale。
This issue was closed because it has been inactive for 14 days since being marked as stale. 当前issue 被标记为stale已有14天,即将关闭。