Skip to content

Issues: OpenBMB/MiniCPM-V

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

88 Open 503 Closed
88 Open 503 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

[BUG] <title>多图单轮对话全量SFT，loss 变为零
#679 opened Nov 25, 2024 by Evan9978
2 tasks done
💡 [REQUEST] - 使用OCR识别场景中，若图片有手写删除的痕迹，是否可以准确理解并屏蔽不需的信息？ question Further information is requested
#678 opened Nov 21, 2024 by evanlin88
提示词“使用OCR识别”，识别结果有：正文/标题的Markdown格式输出，是否可以通过提示词屏蔽？
#677 opened Nov 21, 2024 by evanlin88
2 tasks done
你好，由于 minicpm2.6 的 pretrain ckpt 没有开源，所以这部分评测代码也没有提供，我们的 few-shot 推理方式可以参考 readme 部分的 icl 推理方式。
#676 opened Nov 20, 2024 by ColorDavid
1
推理时，能否直接输出视觉特征？
#675 opened Nov 20, 2024 by kiter-zero
[vllm] - <title> vllm 部署了API接口，示例里边有单张图片的推理，但是多张图片的推理，有参考示例吗？ question Further information is requested
#674 opened Nov 19, 2024 by lh3707
请问sft训练数据中能混入无图数据吗
#673 opened Nov 19, 2024 by 128Ghe980
1
💡 [REQUEST] - <title>预训练相关 question Further information is requested
#672 opened Nov 18, 2024 by zhangzhixun1999
1
Can you provide pre-training files for MiniiCPM-V-2.0 version?
#671 opened Nov 17, 2024 by Leke-G
2 tasks done
微调中断后如何继续训练
#670 opened Nov 16, 2024 by wanghaolan123
1
💡 [REQUEST] - 希望能够实现函数调用的功能，并且提供说明文档，方便第三方写插件 question Further information is requested
#669 opened Nov 15, 2024 by Johnson-yue
3
Accelerator' object has no attribute 'deepspeed_engine_wrapped'
#668 opened Nov 15, 2024 by alanMachineLeraning
RuntimeError: Input type (torch.cuda.HalfTensor) and weight type (torch.HalfTensor) should be the same
#667 opened Nov 15, 2024 by alanMachineLeraning
[BUG] <title> 使用git拉取的DeepSpeed仓库按照命令pip install e . 安装的deepspeed版本号为deepspeed-0.15.4+unknown，后面运行微调命令时，遇到报错，不知道是不是和这个unknown的出现有关系
#666 opened Nov 14, 2024 by xueyuG
2 tasks done
[BUG] <title>为什么对于低分辨率256*256的图像输入的时候，调整max_slice_number = 1/9，对于结果有巨大的差异？
#665 opened Nov 12, 2024 by flawsss
2 tasks done
1
[BUG] <title> Multi turns conversation 's output is not complete!
#663 opened Nov 11, 2024 by FantasticZihao
2 tasks done
5
[BUG] <title>token截断导致训练label被删除
#662 opened Nov 11, 2024 by tong-1989
2 tasks done
2
MiniCPM-V-2.6 promp template的明文格式？
#661 opened Nov 9, 2024 by apepkuss
3
[BUG] 为什么如何调整参数，显存占用都是接近80G？
#660 opened Nov 8, 2024 by DankoZhang
2 tasks done
2
LLM部署MiniCPM-V-2.6并发调用推理报错
#659 opened Nov 7, 2024 by cheng358
2 tasks done
1
[BUG] WebDemo中使用Few Shot功能，如果不传图片无法正常使用
#658 opened Nov 7, 2024 by gongjimin
2 tasks done
[llamacpp] - <title> 为什么llamacpp执行量化模型还要指定一个 f16的 mmproj-model-f16.gguf ? question Further information is requested
#657 opened Nov 6, 2024 by friendmine
1
请问为什么MiniCPM-V-2_6-int4的参数量只有4.76B？
#655 opened Nov 3, 2024 by quanmou
2 tasks done
1
[BUG] MiniCPM-V-2.6, with no image input, answers "As a large language model trained by OpenAI"
#654 opened Oct 28, 2024 by emanuelevivoli
2 tasks done
1
[BUG] <title> Is there a way to output logits?
#653 opened Oct 28, 2024 by zhaowenZhou
2 tasks done
1
ProTip! Adding no:label will show everything without a label.