You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you very much for your excellent work. I have some questions to ask you. I need to implement a specific tag list that includes some more common scenes, objects, attributes, and actions. So may I ask if I need to retrain the label embedding when fine-tuning, because I saw that the you froze the label embedding layer in finetune.py. So, what size of dataset do I need to prepare to achieve the fine-tuning effect.
The text was updated successfully, but these errors were encountered:
Thank you very much for your excellent work. I have some questions to ask you. I need to implement a specific tag list that includes some more common scenes, objects, attributes, and actions. So may I ask if I need to retrain the label embedding when fine-tuning, because I saw that the you froze the label embedding layer in finetune.py. So, what size of dataset do I need to prepare to achieve the fine-tuning effect.
The text was updated successfully, but these errors were encountered: