stepfun-ai
/

GOT-OCR2_0

Image-Text-to-Text

feature-extraction

vision-language

Model card Files Files and versions Community

ucaslcl commited on 29 days ago

Commit

355b4db

•

1 Parent(s): 33eaccc

Update modeling_GOT.py

Files changed (1) hide show

modeling_GOT.py +1 -1

modeling_GOT.py CHANGED Viewed

@@ -237,7 +237,7 @@ class GOTQwenModel(Qwen2Model):
             for image in images:
                 P, C, H, W = image.shape
-                print(image)
                 if P == 1:
                     with torch.set_grad_enabled(False):
                         cnn_feature = vision_tower_high(image)

             for image in images:
                 P, C, H, W = image.shape
+                print(image.shape)
                 if P == 1:
                     with torch.set_grad_enabled(False):
                         cnn_feature = vision_tower_high(image)