Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

开启断点训练后,为什么剩余时间越来越大了,epoch 和max_steps也和断点训练前接不上呢 #3783

Open
evyy1234 opened this issue Apr 7, 2025 · 2 comments

Comments

@evyy1234
Copy link

evyy1234 commented Apr 7, 2025

如图:Image

断点训练参数如下:
NPROC_PER_NODE=2 CUDA_VISIBLE_DEVICES=0,1 swift sft --model_type qwen2_5 --model /opt/db/Qwen2___5-32B-Instruct --resume_from_checkpoint /opt/db/output/Qwen2.5-32B-Instruct/v16-20250402-202739/checkpoint-5393 --dataset data/ap.jsonl --sequence_parallel_size 2 --deepspeed 'zero3'

@evyy1234
Copy link
Author

evyy1234 commented Apr 7, 2025

@Jintao-Huang

@Jintao-Huang
Copy link
Collaborator

--load_args true

观察 num_train_epochs相比第一次训练是否有变化

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants