def get_test_pipeline(self, model, tokenizer, feature_extractor):
image_segmenter = ImageSegmentationPipeline(
model=model, feature_extractor=feature_extractor)
return image_segmenter, [
"./tests/fixtures/tests_samples/COCO/000000039769.png",
"./tests/fixtures/tests_samples/COCO/000000039769.png",
]
def run_pipeline_test(self, model, tokenizer, feature_extractor):
image_segmenter = ImageSegmentationPipeline(model=model, feature_extractor=feature_extractor)
outputs = image_segmenter("./tests/fixtures/tests_samples/COCO/000000039769.png", threshold=0.0)
self.assertEqual(outputs, [{"score": ANY(float), "label": ANY(str), "mask": ANY(str)}] * 12)
import datasets
dataset = datasets.load_dataset("Narsil/image_dummy", "image", split="test")
batch = [
Image.open("./tests/fixtures/tests_samples/COCO/000000039769.png"),
"http://images.cocodataset.org/val2017/000000039769.jpg",
# RGBA
dataset[0]["file"],
# LA
dataset[1]["file"],
# L
dataset[2]["file"],
]
outputs = image_segmenter(batch, threshold=0.0)
self.assertEqual(len(batch), len(outputs))
self.assertEqual(
outputs,
[
[{"score": ANY(float), "label": ANY(str), "mask": ANY(str)}] * 12,
[{"score": ANY(float), "label": ANY(str), "mask": ANY(str)}] * 12,
[{"score": ANY(float), "label": ANY(str), "mask": ANY(str)}] * 12,
[{"score": ANY(float), "label": ANY(str), "mask": ANY(str)}] * 12,
[{"score": ANY(float), "label": ANY(str), "mask": ANY(str)}] * 12,
],
)
def test_small_model_pt(self):
model_id = "hf-internal-testing/tiny-detr-mobilenetsv3-panoptic"
model = AutoModelForImageSegmentation.from_pretrained(model_id)
feature_extractor = AutoFeatureExtractor.from_pretrained(model_id)
image_segmenter = ImageSegmentationPipeline(
model=model, feature_extractor=feature_extractor)
outputs = image_segmenter(
"http://images.cocodataset.org/val2017/000000039769.jpg",
threshold=0.0)
for o in outputs:
# shortening by hashing
o["mask"] = hashimage(o["mask"])
self.assertEqual(
nested_simplify(outputs, decimals=4),
[
{
"score": 0.004,
"label": "LABEL_0",
"mask": "34eecd16bbfb0f476083ef947d81bf66",
},
{
"score": 0.004,
"label": "LABEL_0",
"mask": "34eecd16bbfb0f476083ef947d81bf66",
},
],
)
outputs = image_segmenter(
[
"http://images.cocodataset.org/val2017/000000039769.jpg",
"http://images.cocodataset.org/val2017/000000039769.jpg",
],
threshold=0.0,
)
for output in outputs:
for o in output:
o["mask"] = hashimage(o["mask"])
self.assertEqual(
nested_simplify(outputs, decimals=4),
[
[
{
"score": 0.004,
"label": "LABEL_0",
"mask": "34eecd16bbfb0f476083ef947d81bf66",
},
{
"score": 0.004,
"label": "LABEL_0",
"mask": "34eecd16bbfb0f476083ef947d81bf66",
},
],
[
{
"score": 0.004,
"label": "LABEL_0",
"mask": "34eecd16bbfb0f476083ef947d81bf66",
},
{
"score": 0.004,
"label": "LABEL_0",
"mask": "34eecd16bbfb0f476083ef947d81bf66",
},
],
],
)
def test_small_model_pt(self):
model_id = "mishig/tiny-detr-mobilenetsv3-panoptic"
model = AutoModelForImageSegmentation.from_pretrained(model_id)
feature_extractor = AutoFeatureExtractor.from_pretrained(model_id)
image_segmenter = ImageSegmentationPipeline(
model=model, feature_extractor=feature_extractor)
outputs = image_segmenter(
"http://images.cocodataset.org/val2017/000000039769.jpg",
threshold=0.0)
for o in outputs:
# shortening by hashing
o["mask"] = hashlib.sha1(o["mask"].encode("UTF-8")).hexdigest()
self.assertEqual(
nested_simplify(outputs, decimals=4),
[
{
"score": 0.004,
"label": "LABEL_0",
"mask": "4276f7db4ca2983b2666f7e0c102d8186aed20be",
},
{
"score": 0.004,
"label": "LABEL_0",
"mask": "4276f7db4ca2983b2666f7e0c102d8186aed20be",
},
],
)
outputs = image_segmenter(
[
"http://images.cocodataset.org/val2017/000000039769.jpg",
"http://images.cocodataset.org/val2017/000000039769.jpg",
],
threshold=0.0,
)
for output in outputs:
for o in output:
o["mask"] = hashlib.sha1(o["mask"].encode("UTF-8")).hexdigest()
self.assertEqual(
nested_simplify(outputs, decimals=4),
[
[
{
"score": 0.004,
"label": "LABEL_0",
"mask": "4276f7db4ca2983b2666f7e0c102d8186aed20be",
},
{
"score": 0.004,
"label": "LABEL_0",
"mask": "4276f7db4ca2983b2666f7e0c102d8186aed20be",
},
],
[
{
"score": 0.004,
"label": "LABEL_0",
"mask": "4276f7db4ca2983b2666f7e0c102d8186aed20be",
},
{
"score": 0.004,
"label": "LABEL_0",
"mask": "4276f7db4ca2983b2666f7e0c102d8186aed20be",
},
],
],
)