本文整理汇总了Python中slim.datasets.dataset_utils.image_to_tfexample方法的典型用法代码示例。如果您正苦于以下问题：Python dataset_utils.image_to_tfexample方法的具体用法？Python dataset_utils.image_to_tfexample怎么用？Python dataset_utils.image_to_tfexample使用的例子？那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类slim.datasets.dataset_utils的用法示例。
在下文中一共展示了dataset_utils.image_to_tfexample方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: _convert_dataset
# 需要导入模块: from slim.datasets import dataset_utils [as 别名]
# 或者: from slim.datasets.dataset_utils import image_to_tfexample [as 别名]
def _convert_dataset(split_name, filenames, filename_to_class_id, dataset_dir):
"""Converts the given filenames to a TFRecord dataset.
Args:
split_name: The name of the dataset, either 'train' or 'valid'.
filenames: A list of absolute paths to png images.
filename_to_class_id: A dictionary from filenames (strings) to class ids
(integers).
dataset_dir: The directory where the converted datasets are stored.
"""
print('Converting the {} split.'.format(split_name))
# Train and validation splits are both in the train directory.
if split_name in ['train', 'valid']:
png_directory = os.path.join(dataset_dir, 'mnist_m', 'mnist_m_train')
elif split_name == 'test':
png_directory = os.path.join(dataset_dir, 'mnist_m', 'mnist_m_test')
with tf.Graph().as_default():
image_reader = ImageReader()
with tf.Session('') as sess:
output_filename = _get_output_filename(dataset_dir, split_name)
with tf.python_io.TFRecordWriter(output_filename) as tfrecord_writer:
for filename in filenames:
# Read the filename:
image_data = tf.gfile.FastGFile(
os.path.join(png_directory, filename), 'r').read()
height, width = image_reader.read_image_dims(sess, image_data)
class_id = filename_to_class_id[filename]
example = dataset_utils.image_to_tfexample(image_data, 'png', height,
width, class_id)
tfrecord_writer.write(example.SerializeToString())
sys.stdout.write('\n')
sys.stdout.flush()