def test_save_json():
experiment_id = "test_save_json"
environment.init(experiment_id)
setup_test_environment()
test_dict = {
"model_name": "save_json",
"image_size_height": 128,
"image_size_width": 64,
"num_classes": 3,
"parameters": "test_node",
"flops": "test_flops",
}
if not os.path.exists(environment.EXPERIMENT_DIR):
os.makedirs(environment.EXPERIMENT_DIR)
_save_json(name=test_dict["model_name"],
image_size=(test_dict["image_size_height"],
test_dict["image_size_width"]),
num_classes=test_dict["num_classes"],
node_param_dict=test_dict["parameters"],
node_flops_dict=test_dict["flops"])
output_file = os.path.join(
environment.EXPERIMENT_DIR,
"{}_profile.json".format(test_dict["model_name"]))
with open(output_file, 'r') as fp:
file_data = json.load(fp)
assert os.path.isfile(output_file)
assert test_dict == file_data
def test_build_tfds_object_detection():
environment.setup_test_environment()
# Build TFDS Dataset
config_file = "unit/fixtures/configs/for_build_tfds_object_detection.py"
run(config_file, overwrite=True)
# Check if the builded dataset can be loaded with the same config file
expriment_id = "tfds_object_detection"
train_run(config_file, expriment_id, recreate=True)
# Check if the dataset was build correctly
train_data_num = 3
validation_data_num = 2
config = config_util.load(config_file)
train_dataset = setup_dataset(TFDSObjectDetection,
subset="train",
batch_size=config.BATCH_SIZE,
**config.DATASET.TFDS_KWARGS)
validation_dataset = setup_dataset(TFDSObjectDetection,
subset="validation",
batch_size=config.BATCH_SIZE,
**config.DATASET.TFDS_KWARGS)
assert train_dataset.num_per_epoch == train_data_num
assert validation_dataset.num_per_epoch == validation_data_num
assert train_dataset.num_max_boxes == validation_dataset.num_max_boxes
num_max_boxes = train_dataset.num_max_boxes
for _ in range(train_data_num):
images, labels = train_dataset.feed()
assert isinstance(images, np.ndarray)
assert images.shape[0] == config.BATCH_SIZE
assert images.shape[1] == config.IMAGE_SIZE[0]
assert images.shape[2] == config.IMAGE_SIZE[1]
assert images.shape[3] == 3
assert isinstance(labels, np.ndarray)
assert labels.shape[0] == config.BATCH_SIZE
assert labels.shape[1] == num_max_boxes
assert labels.shape[2] == 5
for _ in range(validation_data_num):
images, labels = validation_dataset.feed()
assert isinstance(images, np.ndarray)
assert images.shape[0] == config.BATCH_SIZE
assert images.shape[1] == config.IMAGE_SIZE[0]
assert images.shape[2] == config.IMAGE_SIZE[1]
assert images.shape[3] == 3
assert isinstance(labels, np.ndarray)
assert labels.shape[0] == config.BATCH_SIZE
assert labels.shape[1] == num_max_boxes
assert labels.shape[2] == 5
def test_export():
config_file = "unit/fixtures/configs/for_export.py"
expriment_id = "test_export"
train_run(config_file, expriment_id, recreate=True, profile_step=7)
setup_test_environment()
run(expriment_id, None, (None, None), [], None)
def test_predict_object_detection():
config_file = "unit/fixtures/configs/for_predict_object_detection.py"
expriment_id = "test_predict_object_detection"
train_run(config_file, expriment_id, recreate=True)
setup_test_environment()
run("unit/fixtures/sample_images", "outputs", expriment_id, None, None, save_images=True)
def test_profile():
config_file = "unit/fixtures/configs/for_profile.py"
expriment_id = "test_profile"
train_run(config_file, expriment_id, recreate=True)
setup_test_environment()
run(expriment_id, None, None, 2, [])
def set_test_environment():
"""Set test environment"""
print("set test environment")
yield environment.setup_test_environment()
# By using a yield statement instead of return, all the code after the yield statement serves as the teardown code:
# See also: https://docs.pytest.org/en/latest/fixture.html#fixture-finalization-executing-teardown-code
environment.teardown_test_environment()
assert isinstance(labels, np.ndarray)
assert labels.shape[0] == batch_size
assert labels.shape[1] == num_max_boxes
assert labels.shape[2] == 5
for _ in range(2):
images, labels = validation_dataset.feed()
# _show_images_with_boxes(images, labels)
assert isinstance(images, np.ndarray)
assert images.shape[0] == batch_size
assert images.shape[1] == image_size[0]
assert images.shape[2] == image_size[1]
assert images.shape[3] == 3
assert isinstance(labels, np.ndarray)
assert labels.shape[0] == batch_size
assert labels.shape[1] == num_max_boxes
assert labels.shape[2] == 5
if __name__ == '__main__':
from blueoil.environment import setup_test_environment
setup_test_environment()
test_delta_mark_classification()
test_delta_mark_classification_has_validation_path()
test_delta_mark_object_detection()
test_delta_mark_object_detection_has_validation_path()
dataset = Dummy(subset="train", batch_size=batch_size)
dataset_iterator = DatasetIterator(dataset)
for i in range(0, 10):
images, labels = next(dataset_iterator)
assert images.shape[0] == batch_size
assert labels.shape[0] == batch_size
def test_dataset_iterator_batch_order():
"""Assert that data given by iterator is same whether enabele_prefetch ture or false."""
batch_size = 8
dataset = Dummy(subset="train", batch_size=batch_size)
dataset_iterator = DatasetIterator(dataset, seed=10, enable_prefetch=False)
prefetch_dataset_iterator = DatasetIterator(dataset, seed=10, enable_prefetch=True)
for i in range(0, 30):
images, labels = next(dataset_iterator)
prefetch_images, prefetch_labels = next(prefetch_dataset_iterator)
assert np.all(images == prefetch_images)
assert np.all(labels == prefetch_labels)
if __name__ == '__main__':
from blueoil import environment
environment.setup_test_environment()
test_dataset_iterator_batch_size()
test_dataset_iterator_batch_order()
