Paddle Multimodal Integration and eXploration, supporting mainstream multi-modal tasks, including end-to-end large-scale multi-modal pretrain models and diffusion model toolbox. Equipped with high performance and flexibility.
AttributeError: 'LlavaConfig' object has no attribute 'pretrain_mm_mlp_adapter'. Did you mean: 'freeze_mm_mlp_adapter'?
pretrain_mm_mlp_adapter似乎是加载projector的路径，我用的模型是hugging face上下载的llava1.5-7b，我尝试手动更改加载projector，但提示了warning：projector0和projector2的weight和bias并没有初始化，最终预测的输出也是乱码。请问我改如何进行推理
wojiaoshihua
commented
2 months ago
AttributeError: 'LlavaConfig' object has no attribute 'pretrain_mm_mlp_adapter'. Did you mean: 'freeze_mm_mlp_adapter'? pretrain_mm_mlp_adapter似乎是加载projector的路径，我用的模型是hugging face上下载的llava1.5-7b，我尝试手动更改加载projector，但提示了warning：projector0和projector2的weight和bias并没有初始化，最终预测的输出也是乱码。请问我改如何进行推理