LinWeizheDragon / FLMR

The huggingface implementation of Fine-grained Late-interaction Multi-modal Retriever.
42 stars 2 forks source link

请问这个模型可以只传入文本,不传入图片吗? #7

Closed jinlong7790 closed 2 months ago

LinWeizheDragon commented 3 months ago

Yes, you can input only a black image, as did in the pertaining process. Or if you want, you can inspect the code and set input_modality to use only text encoder in retrieval - this removes the vision part entirely.