AttributeError: 'numpy.ndarray' object has no attribute 'clone' (when using image coordinates for box prompt)

[barakhurwitz]

Environment Information

• samgeo version: v0.10.1
• Python version: 3.10.12
• Operating System: running in colab

Description

Describe what you were trying to get done.
Tell us what happened, what went wrong, and what you expected to happen.

What I Did

Trying to segment using box prompt
( boxes defined without CRS , i want to use image coordinates )
based on this box prompt example: , :https://samgeo.gishub.org/examples/box_prompts/
trying to run without CRS

boxes_2 = [
        [0, 0, 162, 56],
        [227, 0, 310, 24],  
    ]

sam.predict(boxes=boxes_2 , output="mask.tif", dtype="uint8")

Error:
AttributeError: 'numpy.ndarray' object has no attribute 'clone'

traceback :
/usr/local/lib/python3.10/dist-packages/samgeo/samgeo.py in predict(self, point_coords, point_labels, boxes, point_crs, mask_input, multimask_output, return_logits, output, index, mask_multiplier, dtype, return_results, **kwargs)
583 )
584 else:
--> 585 masks, scores, logits = predictor.predict_torch(
586 point_coords=point_coords,
587 point_labels=point_coords,

/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py in decorate_context(*args, **kwargs)
113 def decorate_context(*args, **kwargs):
114 with ctx_factory():
--> 115 return func(*args, **kwargs)
116
117 return decorate_context

/usr/local/lib/python3.10/dist-packages/segment_anything/predictor.py in predict_torch(self, point_coords, point_labels, boxes, mask_input, multimask_output, return_logits)
220
221 # Embed prompts
--> 222 sparse_embeddings, dense_embeddings = self.model.prompt_encoder(
223 points=points,
224 boxes=boxes,

/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, **kwargs)
1499 or _global_backward_pre_hooks or _global_backward_hooks
1500 or _global_forward_hooks or _global_forward_pre_hooks):
-> 1501 return forward_call(*args, **kwargs)
1502 # Do not call functions when jit is used
1503 full_backward_hooks, non_full_backward_hooks = [], []

/usr/local/lib/python3.10/dist-packages/segment_anything/modeling/prompt_encoder.py in forward(self, points, boxes, masks)
156 sparse_embeddings = torch.cat([sparse_embeddings, point_embeddings], dim=1)
157 if boxes is not None:
--> 158 box_embeddings = self._embed_boxes(boxes)
159 sparse_embeddings = torch.cat([sparse_embeddings, box_embeddings], dim=1)
160

/usr/local/lib/python3.10/dist-packages/segment_anything/modeling/prompt_encoder.py in _embed_boxes(self, boxes)
95 boxes = boxes + 0.5 # Shift to center of pixel
96 coords = boxes.reshape(-1, 2, 2)
---> 97 corner_embedding = self.pe_layer.forward_with_coords(coords, self.input_image_size)
98 corner_embedding[:, 0, :] += self.point_embeddings[2].weight
99 corner_embedding[:, 1, :] += self.point_embeddings[3].weight

/usr/local/lib/python3.10/dist-packages/segment_anything/modeling/prompt_encoder.py in forward_with_coords(self, coords_input, image_size)
209 ) -> torch.Tensor:
210 """Positionally encode points that are not normalized to [0,1]."""
--> 211 coords = coords_input.clone()
212 coords[:, :, 0] = coords[:, :, 0] / image_size[1]
213 coords[:, :, 1] = coords[:, :, 1] / image_size[0]

AttributeError: 'numpy.ndarray' object has no attribute 'clone'