dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update image_processing_deformable_detr.py (#28561) 

* Update image_processing_deformable_detr.py

* Changes after running make fix-copies
This commit is contained in:
Sounak Dey 2024-01-22 16:17:39 +01:00 committed by GitHub
parent e201864bcb
commit a35ea570a8
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 16 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
src/transformers/models/conditional_detr/image_processing_conditional_detr.py
View File

@ -1414,13 +1414,14 @@ class ConditionalDetrImageProcessor(BaseImageProcessor):
boxes = torch.gather(boxes, 1, topk_boxes.unsqueeze(-1).repeat(1, 1, 4))
# and from relative [0, 1] to absolute [0, height] coordinates
if isinstance(target_sizes, List):
img_h = torch.Tensor([i[0] for i in target_sizes])
img_w = torch.Tensor([i[1] for i in target_sizes])
else:
img_h, img_w = target_sizes.unbind(1)
scale_fct = torch.stack([img_w, img_h, img_w, img_h], dim=1).to(boxes.device)
boxes = boxes * scale_fct[:, None, :]
if target_sizes is not None:
if isinstance(target_sizes, List):
img_h = torch.Tensor([i[0] for i in target_sizes])
img_w = torch.Tensor([i[1] for i in target_sizes])
else:
img_h, img_w = target_sizes.unbind(1)
scale_fct = torch.stack([img_w, img_h, img_w, img_h], dim=1).to(boxes.device)
boxes = boxes * scale_fct[:, None, :]
results = []
for s, l, b in zip(scores, labels, boxes):

15
src/transformers/models/deformable_detr/image_processing_deformable_detr.py
View File

@ -1411,13 +1411,14 @@ class DeformableDetrImageProcessor(BaseImageProcessor):
boxes = torch.gather(boxes, 1, topk_boxes.unsqueeze(-1).repeat(1, 1, 4))
# and from relative [0, 1] to absolute [0, height] coordinates
if isinstance(target_sizes, List):
img_h = torch.Tensor([i[0] for i in target_sizes])
img_w = torch.Tensor([i[1] for i in target_sizes])
else:
img_h, img_w = target_sizes.unbind(1)
scale_fct = torch.stack([img_w, img_h, img_w, img_h], dim=1).to(boxes.device)
boxes = boxes * scale_fct[:, None, :]
if target_sizes is not None:
if isinstance(target_sizes, List):
img_h = torch.Tensor([i[0] for i in target_sizes])
img_w = torch.Tensor([i[1] for i in target_sizes])
else:
img_h, img_w = target_sizes.unbind(1)
scale_fct = torch.stack([img_w, img_h, img_w, img_h], dim=1).to(boxes.device)
boxes = boxes * scale_fct[:, None, :]
results = []
for s, l, b in zip(scores, labels, boxes):