Do we need image_queue and text_queue in fine-tuning?

salesforce / ALBEF

Code for ALBEF: a new vision-language pre-training method

BSD 3-Clause "New" or "Revised" License

1.57k stars 199 forks source link

Closed AHEADer closed 2 years ago

AHEADer commented 2 years ago

I've noticed the pretrained weights contain image_queue, text_queue and queue_ptr. Is it recommended to load these 3 parameters for finetuning?

LiJunnan1992 commented 2 years ago

Hi, those are not necessary during fine-tuning.

AHEADer commented 2 years ago

Thanks!