Few shot training with SAA+

[zeddo123]

Hello,

I've been reading the paper on the model and there isn't much mention of Few Shot Learning in it. However, in the code there's an argument k_shot, is this referring to few shot learning? if so could you shed some light on how it is implemented in this context? If not how it might be done it SAA+?

[caoyunkang]

Thank you for your question! Currently, SAA+ does not support few-shot fine-tuning. Fine-tuning is highly dependent on Grounding-DINO, which has not yet released its training/fine-tuning code.

In the meantime, we encourage you to explore our latest work, AdaCLIP, where you can find the training process available in the repository.