Open lgs00 opened 1 year ago
Hello, thank you for your contribution. I meet a question on _line 66 of the file models/spillava.py， image_forward_outs = vision_tower(images,output_hidden_states=True) What is the structure of this vision_tower？
image_forward_outs = vision_tower(images,output_hidden_states=True)
The vision tower is CLIP ViT-H/14.
jshilong
commented
1 year ago
Hello, thank you for your contribution. I meet a question on _line 66 of the file models/spillava.py，
image_forward_outs = vision_tower(images,output_hidden_states=True)What is the structure of this vision_tower？