Skip to content

Issues: modelscope/ms-swift

GRPO (R1) 训练交流群
#3076 opened Feb 12, 2025 by Jintao-Huang
Open 3
ms-swift3 Suggestion Box
#2217 opened Oct 10, 2024 by Jintao-Huang
Open 34
SWIFT Support GRPO Agent Training
#3194 opened Feb 20, 2025 by tastelikefeet
Open 1
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.

Sign up for GitHub

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

379 Open 1,350 Closed
379 Open 1,350 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

GRPO训练LLAVA CUDA Error
#3264 opened Feb 25, 2025 by MarkDeng1
npu zero3 训练自定义模型时，报错Function SumBackward0 returned an invalid gradient at index 0
#3262 opened Feb 25, 2025 by Zane-Qbb
【Bug】DeepSpeed Zero 3 训练 Qwen2.5 32B GRPO Beta=0 时报错
#3261 opened Feb 25, 2025 by Peng-YM
3
训练RegressionTask的一些问题
#3260 opened Feb 25, 2025 by NicholasKX
1
Why is flash attention not supported when fine tuning ovis2
#3258 opened Feb 24, 2025 by mehamednews
为什么在get_model_tokenizer_from_local函数里要将past_key_values添加到model_config.keys_to_ignore_at_inference中啊
#3257 opened Feb 24, 2025 by zsxm1998
1
Can I continue fine-tuning based on the results of the last fine-tuning?
#3253 opened Feb 24, 2025 by safwaqf
1
internvl2.5微调
#3250 opened Feb 24, 2025 by Ranking666
2
How to set ‘max_pixels’ and ‘min_pixels’ for Qwen2.5-VL's processor
#3245 opened Feb 24, 2025 by zhangle09
1
GRPO多机LMDeploy训练warning，loss为0
#3244 opened Feb 24, 2025 by owl-10
设置了seq_cls后，怎么自定义每个label的loss权重？
#3240 opened Feb 24, 2025 by kike-0304
1
grpo 训练deepspeed==0.14.5报错
#3239 opened Feb 24, 2025 by rtz1998
1
请问怎么用python实现类似的多轮对话功能呢？
#3238 opened Feb 24, 2025 by taylover-pei
3
测试使用原始精度模型微调的时候没问题，但是相同配置，使用低精度Int4微调的时候出错了？
#3235 opened Feb 24, 2025 by thinkbig
GRPO LLava 训练报错，无法多卡训练，1卡可以
#3228 opened Feb 23, 2025 by MarkDeng1
请问支持Baichuan-M1-14B-Instruct吗？
#3227 opened Feb 23, 2025 by Leonand6
请问支持soft label的分类任务训练吗
#3224 opened Feb 22, 2025 by waltonfuture
GRPO 4卡A100训练BUG bug Something isn't working
#3223 opened Feb 21, 2025 by MarkDeng1
1
是否支持Megatron-Deepspeed的模型并行？
#3219 opened Feb 21, 2025 by RyanOvO
微调QWen2.5-VL自定义数据集格式支持HTML格式吗？
#3217 opened Feb 21, 2025 by qianyue76
2
请问序列并行是否支持多模态任务 Qwen 2.5 VL
#3215 opened Feb 21, 2025 by lisonsown960321
请问支持 webdataset 作为 qwen2.5VL 的输入么？
#3214 opened Feb 21, 2025 by rover5056
2
使用grpo训练Qwen2.5-VL-7B-Instruct，zero3下混合纯文本和图文数据，训练会卡住 bug Something isn't working
#3213 opened Feb 21, 2025 by Xray112247
如何对deepseek r1做sft和grpo微调
#3211 opened Feb 21, 2025 by ZTurboX
2
关于deepspeed 模式进行ppo训练的问题-》显存占用超额
#3210 opened Feb 21, 2025 by 2043078895
5
ProTip! Adding no:label will show everything without a label.