VOC_2007数据集反序列化

import os
import tensorflow as tf
slim = tf.contrib.slim


def get_dataset(dataset_dir):
    ‘‘‘
    获取数据集
    :param dataset_dir:文件地址
    :return: Dataset
    ‘‘‘
    # 准备参数
    # 第一个参数：数据目录+文件名
    file_pattern = os.path.join(dataset_dir, ‘VOOC_2007_*.tfrecord‘)
    # 第二个参数 reader
    reader = tf.TFRecordReader
    # 第三个参数 decoder
    # 反序列化
    keys_to_features = {
        ‘image/height‘: tf.FixedLenFeature([1], tf.int64),
        ‘image/width‘: tf.FixedLenFeature([1], tf.int64),
        ‘image/channels‘: tf.FixedLenFeature([1], tf.int64),
        ‘image/shape‘: tf.FixedLenFeature([3], tf.int64),
        ‘image/object/bbox/xmin‘: tf.VarLenFeature(dtype=tf.float32),
        ‘image/object/bbox/xmax‘: tf.VarLenFeature(dtype=tf.float32),
        ‘image/object/bbox/ymin‘: tf.VarLenFeature(dtype=tf.float32),
        ‘image/object/bbox/ymax‘: tf.VarLenFeature(dtype=tf.float32),
        ‘image/object/bbox/label‘: tf.VarLenFeature(dtype=tf.int64),
        ‘image/object/bbox/difficult‘: tf.VarLenFeature(dtype=tf.int64),
        ‘image/object/bbox/truncated‘: tf.VarLenFeature(dtype=tf.int64),
        ‘image/format‘: tf.FixedLenFeature((), tf.string, default_value=‘jpeg‘),
        ‘image/encoded‘: tf.FixedLenFeature((), tf.string, default_value=‘‘)
    }
    # 反序列化为高级的形式，用户可以直接使用的形式
    items_to_handlers = {
        ‘image‘: slim.tfexample_decoder.Image(‘image/encoded‘, ‘image/format‘),
        ‘shape‘: slim.tfexample_decoder.Tensor(‘image/shape‘),
        ‘object/bbox‘: slim.tfexample_decoder.BoundingBox(
            [‘ymin‘, ‘xmin‘, ‘ymax‘, ‘xmax‘], ‘image/object/bbox/‘),
        ‘object/labels‘: slim.tfexample_decoder.Tensor(‘image/object/bbox/label‘),
        ‘object/difficult‘: slim.tfexample_decoder.Tensor(‘image/object/bbox/difficult‘),
        ‘object/truncated‘: slim.tfexample_decoder.Tensor(‘image/object/bbox/truncated‘)
    }
    # 构造decoder
    decoder = slim.tfexample_decoder.TFExampleDecoder(keys_to_features, items_to_handlers)

    return slim.dataset.Dataset(
        data_sources = file_pattern,
        reader = reader,
        decoder = decoder,
        num_samples = 9963,
        items_to_descriptions = {
            ‘image‘: ‘A color image of varying height and width‘,
            ‘shape‘: ‘Shape of image‘,
            ‘object/bbox‘: ‘A list of bounding boxes, one per each object‘,
            ‘object/label‘: ‘A list of labels, one per each object‘
        },  # 数据集返回的格式描述字典
        num_classes = 20

    )
--------------------------------------------------------------------------

from datasets import passvalvoc_2007
import tensorflow as tf

slim = tf.contrib.slim

if __name__ == ‘__main__‘:
    # 获取dataset
    dataset = passvalvoc_2007.get_dataset(‘./DATA/TFRecords/‘)
    # 通过provider取出数据
    provider = slim.dataset_data_provider.DatasetDataProvider(
        dataset,
        num_readers = 3  # 读取线程数
    )
    # 通过get方法获取指定名称的数据（是在准备规范数据dataset时高级格式的名称）
    [image, shape, bbox, label, diff, trunc] = provider.get(
        [‘image‘, ‘shape‘, ‘object/bbox‘, ‘object/labels‘, ‘object/difficult‘, ‘object/truncated‘]
    )
    print(image, shape, bbox, label, diff, trunc)

VOC_2007数据集反序列化

原文：https://www.cnblogs.com/yuganwj/p/13721209.html