#!/usr/bin/env python
# ******************************************************************************
# Copyright 2024 Brainchip Holdings Ltd.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ******************************************************************************
"""
Load COCO dataset
"""
__all__ = ["get_coco_dataset"]
import os
try:
import tensorflow_datasets as tfds
except ImportError:
tfds = None
from ..data_utils import remove_empty_objects, get_dataset_length
[docs]
def get_coco_dataset(data_path, training=False):
""" Loads coco dataset and builds a tf.dataset out of it.
Args:
data_path (str): path to the folder containing coco tfrecords.
training (bool, optional): True to retrieve training data,
False for validation. Defaults to False.
Returns:
tf.dataset, list, int: the requested dataset (train or validation), labels and the dataset
size.
"""
assert tfds is not None, "To load coco dataset, tensorflow-datasets module must\
be installed."
write_dir = os.path.join(data_path, 'tfds', 'data')
download_and_prepare_kwargs = {
'download_dir': os.path.join(write_dir, 'downloaded'),
'download_config': tfds.download.DownloadConfig(manual_dir=data_path)
}
split = 'train' if training else 'validation'
dataset, infos = tfds.load(
'coco/2017',
data_dir=write_dir,
split=split,
shuffle_files=training,
download=True,
download_and_prepare_kwargs=download_and_prepare_kwargs,
with_info=True
)
dataset = dataset.filter(remove_empty_objects)
labels = infos.features['objects']['label'].names
dataset_length = get_dataset_length(dataset)
return dataset, labels, dataset_length