Jason

the same problem. the generated pictures are the same as input picture. how to fix it? @Hangz-nju-cuhk

the same error with you ,what's worese,I have no setting icon.

do you solve it now?

@LZHgrla Thanks, it works. I have a similar question in convert it to llava format. This is my code `python ./xtuner/configs/llava/llama3_8b_instruct_clip_vit_large_p14_336/convert_xtuner_weights_to_llava.py --text_model_id ./llama3_llava_pth/merge --vision_model_id ${vit} --projector_weight llama3_llava_pth/hf/projector/model.safetensors --save_path ./llama3_llava_pth/LLava_format` This...

I also think so at first. but when I load the LLava_format in this way. ``` from llava.model.builder import load_pretrained_model from llava.mm_utils import get_model_name_from_path from llava.eval.run_llava import eval_model model_path =...

> @Jason8Kang Try this script? > > https://huggingface.co/xtuner/llava-llama-3-8b-v1_1-hf/discussions/1 thanks you, it works in your script. but it still has the warning as follows. do you know the meaning of this...

遇到同样问题

懂了,多谢解答