# Copyright (C) 2020 Intel Corporation
#
# SPDX-License-Identifier: MIT
from datumaro.util.tf_util import import_tf
import_tf() # prevent TF loading and potential interpeter crash
import accuracy_checker.representation as ac
import datumaro.components.extractor as dm
from datumaro.util.annotation_util import softmax
def import_predictions(predictions):
# Convert Accuracy checker predictions to Datumaro annotations
anns = []
for pred in predictions:
anns.extend(import_prediction(pred))
return anns
def import_prediction(pred):
if isinstance(pred, ac.ClassificationPrediction):
scores = softmax(pred.scores)
return (dm.Label(label_id, attributes={'score': float(score)})
for label_id, score in enumerate(scores))
elif isinstance(pred, ac.ArgMaxClassificationPrediction):
return (dm.Label(int(pred.label)), )
from datumaro.components.annotation import (
AnnotationType,
Bbox,
Label,
LabelCategories,
)
from datumaro.components.extractor import (
CategoriesInfo,
DatasetItem,
IExtractor,
)
from datumaro.components.media import ByteImage
from datumaro.util.tf_util import import_tf
try:
tf = import_tf()
import tensorflow_datasets as tfds
except ImportError:
log.debug("Unable to import TensorFlow or TensorFlow Datasets. " \
"Dataset downloading via TFDS is disabled.")
TFDS_EXTRACTOR_AVAILABLE = False
else:
TFDS_EXTRACTOR_AVAILABLE = True
@frozen
class TfdsDatasetMetadata:
default_converter_name: str
@frozen
# SPDX-License-Identifier: MIT
from tempfile import TemporaryDirectory
from pyunpack import Archive
from cvat.apps.dataset_manager.bindings import (CvatTaskDataExtractor,
import_dm_annotations)
from cvat.apps.dataset_manager.util import make_zip_archive
from datumaro.components.project import Dataset
from .registry import dm_env, exporter, importer
from datumaro.util.tf_util import import_tf
try:
import_tf()
tf_available = True
except ImportError:
tf_available = False
@exporter(name='TFRecord', ext='ZIP', version='1.0', enabled=tf_available)
def _export(dst_file, task_data, save_images=False):
extractor = CvatTaskDataExtractor(task_data, include_images=save_images)
extractor = Dataset.from_extractors(extractor) # apply lazy transforms
with TemporaryDirectory() as temp_dir:
converter = dm_env.make_converter('tf_detection_api',
save_images=save_images)
converter(extractor, save_dir=temp_dir)
make_zip_archive(temp_dir, dst_file)