Skip to content

Issues: modelscope/ms-swift

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

218 Open 1,092 Closed
218 Open 1,092 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

关于swift web-ui启动后微调AI并使用本地数据库
#2472 opened Nov 18, 2024 by ivandgetic
For multimodal models like Qwen-VL or Qwen-Audio, how to finetune projector with full parameters and finetune language models with LoRA simultaneously?
#2471 opened Nov 18, 2024 by jymh
Add Multimodal Input Support (Image, Audio, Video) to App-UI in MS-Swift Library
#2469 opened Nov 18, 2024 by SushantGautam
Is there option for fine-tuning only bias parameters ?
#2468 opened Nov 18, 2024 by thangld201
请问如果我在sft的参数里添加warmup_ratio, 然后用deepspeed去训练，这个warmup_ratio参数会生效么？
#2467 opened Nov 18, 2024 by samaritan1998
Evaluation Crashed
#2466 opened Nov 18, 2024 by mjkim-5090
UnboundLocalError: local variable 'linear_cls' referenced before assignment
#2465 opened Nov 18, 2024 by cryrookie
ValueError: Image features and image tokens do not match: tokens: 5589, features 5805
#2460 opened Nov 16, 2024 by Gary-code
Checkpoint model error
#2459 opened Nov 15, 2024 by BurakYildizak
1
qwen2-vl使用lmdeploy推理报错。 bug Something isn't working
#2457 opened Nov 15, 2024 by bonre
1
使用streaming训练在eval会卡住 bug Something isn't working
#2455 opened Nov 15, 2024 by 1215thebqtic
Qwen2-vl-7B 怎么设置仅微调视觉模型？
#2454 opened Nov 15, 2024 by thesby
2
Error with Qwen2VL When Images Are Missing
#2452 opened Nov 14, 2024 by LukeForeverYoung
2
Regarding the finetune speed different between different swift version
#2451 opened Nov 14, 2024 by Yuuuuuuuuuuuuuuuuuummy
2
可以对现有的checkpoint 追加数据集乾增量训练吗？
#2450 opened Nov 14, 2024 by gracelinlin
1
train_loss与正常loss对不上
#2448 opened Nov 13, 2024 by zhaoyangwei123
docker部署之后，输入swift web-ui，显示无此指令，之前没遇到过这问题，这该怎么解决？
#2447 opened Nov 13, 2024 by Fanxhion
3
NPU上是否支持Qwen2-vl的sft微调？
#2446 opened Nov 13, 2024 by zzzengzhe
vllm 推理微调qwen2_vl_7b异常
#2445 opened Nov 13, 2024 by Gpandaa
多机多卡的情况下，微调会被卡住
#2443 opened Nov 13, 2024 by boynicholas
1
【没有llama3.2-11b-vision最佳实践】
#2442 opened Nov 13, 2024 by LixiangHello
3
是否支持超长文本训练？
#2441 opened Nov 13, 2024 by charliedream1
When deploying Qwen-2VL using Swift, the generated results encounter an early termination issue.
#2440 opened Nov 13, 2024 by gt-liuzijie01
3
微调的命令行启动方式是什么？看文档只有swift web-ui启动，但是链接只有72小时有效期，永久的链接还得部署gradio
#2439 opened Nov 13, 2024 by Fanxhion
1
How to deploy a trained reward model?
#2437 opened Nov 12, 2024 by guru4elephant
ProTip! Adding no:label will show everything without a label.