TF學習之DeepLabv3+代碼閱讀2（dataset）

DeepLabv3+代碼閱讀之data_generator.py

一、Dataset

1. __init__

class Dataset(object):
  """Represents input dataset for deeplab model."""

  def __init__(self,
               dataset_name,# Dataset name
               split_name,# A train/val Split name
               dataset_dir,# The directory of the dataset sources
               batch_size,# Batch size
               crop_size,# The size used to crop the image and label
               min_resize_value=None,# Desired size of the smaller image side
               max_resize_value=None,# Maximum allowed size of the larger image side
               resize_factor=None,# Resized dimensions are multiple of factor plus one
               min_scale_factor=1.,# Minimum scale factor value
               max_scale_factor=1.,# Maximum scale factor value
               scale_factor_step_size=0,
               model_variant=None,# Model variant (string) for choosing how to mean-subtract the images.
        						  # See feature_extractor.network_map for supported model variants.
               num_readers=1,# Number of readers for data provider
               is_training=False,# Boolean, if dataset is for training or not
               should_shuffle=False,# Boolean, if should shuffle the input data
               should_repeat=False):# Boolean, if should repeat the input data

    if dataset_name not in _DATASETS_INFORMATION:
      raise ValueError('The specified dataset is not supported yet.')
    self.dataset_name = dataset_name

    splits_to_sizes = _DATASETS_INFORMATION[dataset_name].splits_to_sizes# 訓練測試集對應的圖片數量

    if split_name not in splits_to_sizes:
      raise ValueError('data split name %s not recognized' % split_name)

    if model_variant is None:# 使用的模型，如xception_65
      tf.logging.warning('Please specify a model_variant. See '
                         'feature_extractor.network_map for supported model '
                         'variants.')

    self.split_name = split_name
    self.dataset_dir = dataset_dir
    self.batch_size = batch_size
    self.crop_size = crop_size
    self.min_resize_value = min_resize_value
    self.max_resize_value = max_resize_value
    self.resize_factor = resize_factor
    self.min_scale_factor = min_scale_factor
    self.max_scale_factor = max_scale_factor
    self.scale_factor_step_size = scale_factor_step_size
    self.model_variant = model_variant
    self.num_readers = num_readers
    self.is_training = is_training
    self.should_shuffle = should_shuffle
    self.should_repeat = should_repeat

    self.num_of_classes = _DATASETS_INFORMATION[self.dataset_name].num_classes# 類別數
    self.ignore_label = _DATASETS_INFORMATION[self.dataset_name].ignore_label# 忽略的標記

2. get_one_shot_iterator

得到一個迭代數據集一次的迭代器
返回：
	tf.data.Iterator類型迭代器

def get_one_shot_iterator(self):
    """Gets an iterator that iterates across the dataset once.

    Returns:
      An iterator of type tf.data.Iterator.
    """

    files = self._get_all_files()# 得到數據集文件（tfreord）的文件路徑列表
	# tf.data.TFRecordDataset， 從一個或多個TFRecord文件生成數據集，map()對數據做自定義的transform
    dataset = (
        tf.data.TFRecordDataset(files, num_parallel_reads=self.num_readers)
        .map(self._parse_function, num_parallel_calls=self.num_readers)# 解析得到圖片、標記等信息
        .map(self._preprocess_image, num_parallel_calls=self.num_readers))

    if self.should_shuffle:# 是否需要打亂順序
      dataset = dataset.shuffle(buffer_size=100)# 從數據集裏選100個組成buffer，每次讀數據從buffer裏隨機取一個，
      											# 再從剩下的裏邊補一個到buffer。最完美的shuffle則是buffer中的個數
      											# 大於等於整個數據集所含的圖片個數。

    if self.should_repeat:
      dataset = dataset.repeat()  # Repeat forever for training.
    else:
      dataset = dataset.repeat(1) # 對於測試則只需要重複一遍數據集即可

    dataset = dataset.batch(self.batch_size).prefetch(self.batch_size)
    # .batch()形成batch，.prefetch()，將生成數據的時間和使用數據的時間分離開,減少空閒時間
    return dataset.make_one_shot_iterator()# 創建一個列舉數據集中元素的迭代器，最新版本tf中已被廢棄

3. _get_all_files

得到所有文件讀取的位置
返回：
	輸入文件的list

def _get_all_files(self):
  file_pattern = _FILE_PATTERN
  file_pattern = os.path.join(self.dataset_dir,
                              file_pattern % self.split_name)
  return tf.gfile.Glob(file_pattern)# tf.gfile.Glob()返回符合模式的文件路徑列表
  # 例如：file_pattern='tensorflow/models/research/deeplab/datasets/pascal_voc_seg/tfrecord/train-*'

其中

_FILE_PATTERN = '%s-*'

4. _parse_function

解析example proto
參數：
	tf.Example格式的Proto
返回：
	字典，包含解析後的圖片、標籤、尺寸、文件名
備註：
	當前僅支持jpeg和png格式的圖片

def _parse_function(self, example_proto):
    # Currently only supports jpeg and png.
    # Need to use this logic because the shape is not known for tf.image.decode_image and we rely on 
    # this info to extend label if necessary.
    def _decode_image(content, channels):
      return tf.cond(
          tf.image.is_jpeg(content),
          lambda: tf.image.decode_jpeg(content, channels),
          lambda: tf.image.decode_png(content, channels))

    features = {
        'image/encoded':
            tf.FixedLenFeature((), tf.string, default_value=''),
        'image/filename':
            tf.FixedLenFeature((), tf.string, default_value=''),
        'image/format':
            tf.FixedLenFeature((), tf.string, default_value='jpeg'),
        'image/height':
            tf.FixedLenFeature((), tf.int64, default_value=0),
        'image/width':
            tf.FixedLenFeature((), tf.int64, default_value=0),
        'image/segmentation/class/encoded':
            tf.FixedLenFeature((), tf.string, default_value=''),
        'image/segmentation/class/format':
            tf.FixedLenFeature((), tf.string, default_value='png'),
    }

    parsed_features = tf.parse_single_example(example_proto, features)

    image = _decode_image(parsed_features['image/encoded'], channels=3)

    label = None
    if self.split_name != common.TEST_SET:
      label = _decode_image(
          parsed_features['image/segmentation/class/encoded'], channels=1)

    image_name = parsed_features['image/filename']
    if image_name is None:
      image_name = tf.constant('')

    sample = {
        common.IMAGE: image,
        common.IMAGE_NAME: image_name,
        common.HEIGHT: parsed_features['image/height'],
        common.WIDTH: parsed_features['image/width'],
    }

    if label is not None:
      if label.get_shape().ndims == 2:
        label = tf.expand_dims(label, 2)
      elif label.get_shape().ndims == 3 and label.shape.dims[2] == 1:
        pass
      else:
        raise ValueError('Input label shape must be [height, width], or '
                         '[height, width, 1].')

      label.set_shape([None, None, 1])

      sample[common.LABELS_CLASS] = label

    return sample

5. _preprocess_image

對圖片和標記做處理
參數：
	包含圖片和標籤的sample
返回：
	處理後的sample

def _preprocess_image(self, sample):

    image = sample[common.IMAGE]
    label = sample[common.LABELS_CLASS]

    original_image, image, label = input_preprocess.preprocess_image_and_label(
        image=image,
        label=label,
        crop_height=self.crop_size[0],
        crop_width=self.crop_size[1],
        min_resize_value=self.min_resize_value,
        max_resize_value=self.max_resize_value,
        resize_factor=self.resize_factor,
        min_scale_factor=self.min_scale_factor,
        max_scale_factor=self.max_scale_factor,
        scale_factor_step_size=self.scale_factor_step_size,
        ignore_label=self.ignore_label,
        is_training=self.is_training,
        model_variant=self.model_variant)

    sample[common.IMAGE] = image# 將sample裏的image更新爲處理過的

    if not self.is_training:
      # Original image is only used during visualization.
      sample[common.ORIGINAL_IMAGE] = original_image

    if label is not None:
      sample[common.LABEL] = label

    # Remove common.LABEL_CLASS key in the sample since it is only used to
    # derive label and not used in training and evaluation.
    sample.pop(common.LABELS_CLASS, None)

    return sample

二、_DATASETS_INFORMATION

數據集的信息，包含有cityscapes等，使用自己的數據集也要補充在這裏。

_DATASETS_INFORMATION = {
    'cityscapes': _CITYSCAPES_INFORMATION,
    'pascal_voc_seg': _PASCAL_VOC_SEG_INFORMATION,
    'ade20k': _ADE20K_INFORMATION,
}

pascal_voc_seg數據集對應的INFORMATION，其中包含了訓練、測試集圖片數量，類別數等信息。由DatasetDescriptor類定義。

_PASCAL_VOC_SEG_INFORMATION = DatasetDescriptor(
    splits_to_sizes={
        'train': 1464,
        'train_aug': 10582,
        'trainval': 2913,
        'val': 1449,
    },
    num_classes=21,
    ignore_label=255,
)

三、DatasetDescriptor

DatasetDescriptor是用來描述數據集性質的元組。使用collections.namedtuple定義，包含數據集的各種信息。

# Named tuple to describe the dataset properties.
DatasetDescriptor = collections.namedtuple(
    'DatasetDescriptor',
    [
        'splits_to_sizes',  # Splits of the dataset into training, val and test.
        'num_classes',  # Number of semantic classes, including the
                        # background class (if exists). For example, there
                        # are 20 foreground classes + 1 background class in
                        # the PASCAL VOC 2012 dataset. Thus, we set
                        # num_classes=21.
        'ignore_label',  # Ignore label value.
    ])