Charimanhua comments

Results 7 comments of


                                            Charimanhua

TypeError: Qwen2Attention.forward() missing 1 required positional argument: 'attention_mask'

> I wanna know if you solve this problem because I encounter the same problem with you even if the model I use is different with you yes I have...

cannot import name 'Gemma2FlashAttention2' from 'transformers.models.gemma2.modeling_gemma2'

> Hello, [@mxchinegod](https://github.com/mxchinegod). Thank you for pointing this issue! We've released the latest version 1.3.4 on PyPi, where this issue has been fixed. updated FlagEmbedding==1.3.4, still have this error.

使用下列代码，如何加载自己训练的 'ViT-L-14'模型指定路径的.pt文件。（比如加载Epoch为2的./data/model/epoch2.pt模型）

请问你们有遇到 ModuleNotFoundError: No module named 'flash_attn.flash_attention' 报错么？请问你们是如何解决的？谢谢！

Inference and fine-tuning support for GOT-OCR2.

微调后，调用微调后的模型报错： ![401730889265_ pic](https://github.com/user-attachments/assets/e86640ea-da3f-49e3-b85e-b8aef8d77071) 如何解决？模型内容如下： ![411730889276_ pic](https://github.com/user-attachments/assets/98aafd05-e359-4ae6-b87a-1dae77bcf595)

Inference and fine-tuning support for GOT-OCR2.

请问merge lora应该在哪一步操作呀？我不是很懂，谢谢！

ms-swift3 Suggestion Box

请问在微调got-ocr2.0后，使用保存的微调模型时，报错model_type的问题，该如何解决？ ![image](https://github.com/user-attachments/assets/56bd84aa-8576-406a-8a63-d759ba6136d0) ![image](https://github.com/user-attachments/assets/2f410c84-f416-462e-933d-9089ab47bc9f)

ms-swift3 Suggestion Box

> > 请问在微调got-ocr2.0后，使用保存的微调模型时，报错model_type的问题，该如何解决？ > > 你需要merge lora. 才会有config.json文件已解决，谢谢！ cd 进微调模型目录下，执行： swift merge-lora --ckpt_dir xxx