排气扇 comments

Results 4 comments of


                                            排气扇

在运行微调脚本报错 torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 1) local_rank: 0

![image](https://github.com/FlagOpen/FlagEmbedding/assets/93046554/57268062-65eb-44b0-82d8-159e799eac09) 作者你好，请问是这样用bash来运行吗？我这样还是报了和你一样的错误。 > @MrRexy-Ling ，就是把微调训练的指令放到 shell脚本里，用bash来启动 run.py, 因为需要开启多个进程，所以一般用 .sh文件来执行模型的训练启动，截图上面就是我的 run.sh的内容，

在运行微调脚本报错 torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 1) local_rank: 0

![image](https://github.com/FlagOpen/FlagEmbedding/assets/93046554/d337fa7d-c6a2-4817-80ee-6a1a9fe49563)这是我的运行指令。 > ![image](https://private-user-images.githubusercontent.com/93046554/299675225-57268062-65eb-44b0-82d8-159e799eac09.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.7SBx95AaFB-O2iQSS-HcwUNecGKPe3DpamhPY1E-MWo) 作者你好，请问是这样用bash来运行吗？我这样还是报了和你一样的错误。 > > > @MrRexy-Ling ，就是把微调训练的指令放到 shell脚本里，用bash来启动 run.py, 因为需要开启多个进程，所以一般用 .sh文件来执行模型的训练启动，截图上面就是我的 run.sh的内容，

Is there a bug in gen_model_answer_baseline.py?

![微信图片_20240410141349](https://github.com/FasterDecoding/Medusa/assets/93046554/2fb0a24f-cbd5-4f81-bcec-f180451dc45a)

OSError

Thank you for your reply!I have fixed that problem.Can you take a look at the question I asked here?：https://github.com/FasterDecoding/Medusa/issues/45