def test_can_save_and_load_with_multiple_labels(self):
source_dataset = Dataset.from_iterable(
[
DatasetItem(id='1',
subset='train',
annotations=[Label(1), Label(3)]),
DatasetItem(id='2',
subset='train',
image=np.zeros((8, 6, 3)),
annotations=[Label(0)]),
DatasetItem(
id='3',
subset='train',
image=np.zeros((2, 8, 3)),
),
],
categories={
AnnotationType.label:
LabelCategories.from_iterable('label_' + str(label)
for label in range(10)),
})
with TestDir() as test_dir:
ImagenetTxtConverter.convert(source_dataset,
test_dir,
save_images=True)
parsed_dataset = ImagenetTxtImporter()(test_dir).make_dataset()
compare_datasets(self,
source_dataset,
parsed_dataset,
require_images=True)
def test_can_save_and_load(self):
source_dataset = Dataset.from_iterable([
DatasetItem(id='1', subset='train',
annotations=[Label(0)]
),
DatasetItem(id='2', subset='train',
annotations=[Label(0)]
),
DatasetItem(id='3', subset='train', image=np.zeros((8, 8, 3)),
annotations=[Label(0)]
),
DatasetItem(id='4', subset='train',
annotations=[Label(1)]
),
DatasetItem(id='5', subset='train', image=np.zeros((4, 8, 3)),
annotations=[Label(1)]
),
DatasetItem(id='6', subset='train',
annotations=[Label(5)]
),
], categories={
AnnotationType.label: LabelCategories.from_iterable(
'label_' + str(label) for label in range(10)),
})
with TestDir() as test_dir:
ImagenetTxtConverter.convert(source_dataset, test_dir,
save_images=True)
parsed_dataset = Dataset.import_from(test_dir, 'imagenet_txt')
compare_datasets(self, source_dataset, parsed_dataset,
require_images=True)
def test_can_save_and_load_with_no_save_images(self):
source_dataset = Dataset.from_iterable([
DatasetItem(id='1', subset='train', annotations=[Label(0)]),
],
categories=['label_0'])
with TestDir() as test_dir:
ImagenetTxtConverter.convert(source_dataset,
test_dir,
save_images=False)
parsed_dataset = Dataset.import_from(test_dir, 'imagenet_txt')
compare_datasets(self, source_dataset, parsed_dataset)
def test_can_save_dataset_with_no_subsets(self):
source_dataset = Dataset.from_iterable([
DatasetItem(id='a/b/c', image=np.zeros((8, 4, 3)),
annotations=[Label(1)]
),
], categories={
AnnotationType.label: LabelCategories.from_iterable(
'label_' + str(label) for label in range(10)),
})
with TestDir() as test_dir:
ImagenetTxtConverter.convert(source_dataset, test_dir,
save_images=True)
parsed_dataset = ImagenetTxtImporter()(test_dir).make_dataset()
compare_datasets(self, source_dataset, parsed_dataset,
require_images=True)
def test_dataset_with_save_dataset_meta_file(self):
source_dataset = Dataset.from_iterable(
[
DatasetItem(id='1', subset='train', annotations=[Label(0)]),
DatasetItem(id='2', subset='train', annotations=[Label(1)]),
],
categories=['label_0', 'label_1'])
with TestDir() as test_dir:
ImagenetTxtConverter.convert(source_dataset,
test_dir,
save_images=False,
save_dataset_meta=True)
parsed_dataset = Dataset.import_from(test_dir, 'imagenet_txt')
self.assertTrue(osp.isfile(osp.join(test_dir,
'dataset_meta.json')))
compare_datasets(self, source_dataset, parsed_dataset)
def test_can_save_and_load_image_with_arbitrary_extension(self):
dataset = Dataset.from_iterable([
DatasetItem(id='a/1',
image=Image(path='a/1.JPEG', data=np.zeros(
(4, 3, 3)))),
DatasetItem(id='b/c/d/2',
image=Image(path='b/c/d/2.bmp',
data=np.zeros((3, 4, 3)))),
],
categories=[])
with TestDir() as test_dir:
ImagenetTxtConverter.convert(dataset, test_dir, save_images=True)
parsed_dataset = Dataset.import_from(test_dir, 'imagenet_txt')
compare_datasets(self,
dataset,
parsed_dataset,
require_images=True)
def test_can_save_dataset_with_cyrillic_and_spaces_in_filename(self):
dataset = Dataset.from_iterable(
[
DatasetItem(id="кириллица с пробелом",
image=np.ones((8, 8, 3)),
annotations=[Label(0), Label(1)]),
],
categories={
AnnotationType.label:
LabelCategories.from_iterable('label_' + str(label)
for label in range(2)),
})
with TestDir() as test_dir:
ImagenetTxtConverter.convert(dataset, test_dir, save_images=True)
parsed_dataset = Dataset.import_from(test_dir, 'imagenet_txt')
compare_datasets(self,
dataset,
parsed_dataset,
require_images=True)