Skip to content

Issues: modelscope/ms-swift

GRPO (R1) 训练交流群
#3076 opened Feb 12, 2025 by Jintao-Huang
Open 5
Megatron-SWIFT训练交流群
#3604 opened Mar 21, 2025 by Jintao-Huang
Open 2
ms-swift3 Suggestion Box
#2217 opened Oct 10, 2024 by Jintao-Huang
Open 41
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

561 Open 1,541 Closed
561 Open 1,541 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

Do swift support these type of data for training multimodal reward model（having value head）
#3888 opened Apr 15, 2025 by zhang123434
grpo 训练卡住
#3887 opened Apr 15, 2025 by zhilinwang1
AttributeError: 'NoneType' object has no attribute 'shape'
#3885 opened Apr 15, 2025 by AntonioSu
微调qwen2.5-vl做点检测的grounding，数据集应该是什么形式
#3883 opened Apr 15, 2025 by jjjjjjj2020
Is there a specific method for training GRPO using Qwen2.5-VL-3B-Instruct with LoRA?
#3882 opened Apr 15, 2025 by sms-s
merge_lora.sh出错
#3881 opened Apr 15, 2025 by reneliury
1
DPO训练log打印日志:logits/chosen和logits/rejected完全一样
#3880 opened Apr 15, 2025 by qq941134965
There seems not to be a single sample in your epoch_iterator
#3878 opened Apr 15, 2025 by Yuccaaa
grpo log bug
#3877 opened Apr 15, 2025 by Evilxya
GRPO 训练100 steps后性能骤降，请问是什么原因
#3876 opened Apr 15, 2025 by xxzhang0927
grpo训练qwen2.5 7B 100steps后性能直线下降
#3875 opened Apr 15, 2025 by xxzhang0927
VAPO支持计划 enhancement New feature or request
#3872 opened Apr 14, 2025 by DogeWatch
grpo训练32b模型OOM
#3871 opened Apr 14, 2025 by zhilinwang1
3
请问一下会支持kimi-vl的训练吗
#3869 opened Apr 14, 2025 by youweihao-tal
1
meet error when quantizing Qwen2.5vl-72B with multi-gpus
#3867 opened Apr 14, 2025 by sys-reasoner
3
GRPO训练报错：Fatal Python error: none_dealloc: deallocating None: bug likely caused by a refcount error in a C extension
#3864 opened Apr 14, 2025 by winni0
1
[WARNING:swift] No training was carried out, which may be due to the dataset being too small or incorrect usage of resume_from_checkpoint.
#3863 opened Apr 13, 2025 by Henchen99
1
Qwen2.5-vl 怎么训练 DPO 模型？
#3861 opened Apr 13, 2025 by thesby
4
AssertionError: quant_method: bnb, quantized model and does not support merge-lora.
#3859 opened Apr 12, 2025 by cahya-wirawan
多机多卡zero3 lora微调后 merge读取时 报错safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer
#3854 opened Apr 12, 2025 by tytcc
2
GRPO Example script results
#3852 opened Apr 12, 2025 by Zzsf11
单张4090对minicpmV2.6进行视频问答微调总是中途OOM
#3849 opened Apr 12, 2025 by zhuqh19
1
Meet GPU OutOfMemory in GRPO training
#3848 opened Apr 12, 2025 by PluseLin
8
GRPO 算法如果设置 reward_model 而不是--reward_funcs ，reward模型和 model都加载到一张卡里去了
#3843 opened Apr 11, 2025 by wellhowtosay
grpo TypeError: CosineReward.__call__() missing 1 required positional argument: 'solution'
#3840 opened Apr 11, 2025 by kanqgg
2
ProTip! Exclude everything labeled bug with -label:bug.