SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

[SAM] Fixes pipeline and adds a dummy pipeline test (#23684)

Browse Source 
* add a dummy pipeline test

* change test name
This commit is contained in:
Younes Belkada
2023-05-23 17:36:49 +02:00
committed by GitHub
parent 71a5ed3433
commit 42baa58f90
2 changed files with 8 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
src/transformers/models/sam/image_processing_sam.py
View File

@@ -934,7 +934,7 @@ def _generate_crop_boxes(
cropped_images, point_grid_per_crop = _generate_crop_images( cropped_images, point_grid_per_crop = _generate_crop_images(
crop_boxes, image, points_grid, layer_idxs, target_size, original_size crop_boxes, image, points_grid, layer_idxs, target_size, original_size
) )
crop_boxes = np.array(crop_boxes)
crop_boxes = crop_boxes.astype(np.float32) crop_boxes = crop_boxes.astype(np.float32)
points_per_crop = np.array([point_grid_per_crop]) points_per_crop = np.array([point_grid_per_crop])
points_per_crop = np.transpose(points_per_crop, axes=(0, 2, 1, 3)) points_per_crop = np.transpose(points_per_crop, axes=(0, 2, 1, 3))

8
tests/models/sam/test_modeling_sam.py
View File

@@ -20,7 +20,7 @@ import unittest
import requests import requests
from transformers import SamConfig, SamMaskDecoderConfig, SamPromptEncoderConfig, SamVisionConfig from transformers import SamConfig, SamMaskDecoderConfig, SamPromptEncoderConfig, SamVisionConfig, pipeline
from transformers.testing_utils import require_torch, slow, torch_device from transformers.testing_utils import require_torch, slow, torch_device
from transformers.utils import is_torch_available, is_vision_available from transformers.utils import is_torch_available, is_vision_available
@@ -751,3 +751,9 @@ class SamModelIntegrationTest(unittest.TestCase):
iou_scores = outputs.iou_scores.cpu() iou_scores = outputs.iou_scores.cpu()
self.assertTrue(iou_scores.shape == (1, 3, 3)) self.assertTrue(iou_scores.shape == (1, 3, 3))
torch.testing.assert_allclose(iou_scores, EXPECTED_IOU, atol=1e-4, rtol=1e-4) torch.testing.assert_allclose(iou_scores, EXPECTED_IOU, atol=1e-4, rtol=1e-4)
def test_dummy_pipeline_generation(self):
generator = pipeline("mask-generation", model="facebook/sam-vit-base", device=torch_device)
raw_image = prepare_image()
_ = generator(raw_image, points_per_batch=64)