def get_split(dataset_dir,file_pattern=None, reader=None):
"""Gets a dataset tuple with instructions for reading flowers.
Args:
split_name: A train/validation split name.
dataset_dir: The base directory of the dataset sources.
file_pattern: The file pattern to use when matching the dataset sources.
It is assumed that the pattern contains a '%s' string so that the split
name can be inserted.
reader: The TensorFlow reader type.
Returns:
A `Dataset` namedtuple.
Raises:
ValueError: if `split_name` is not a valid train/validation split.
"""
split_name = dataset_dir.split("_")[-1].split(".")[0]
if split_name not in SPLITS_TO_SIZES:
raise ValueError('split name %s was not recognized.' % split_name)
file_pattern = dataset_dir
# Allowing None in the signature so that dataset_factory can use the default.
if reader is None:
reader = tf.TFRecordReader
keys_to_features = {
'image/encoded': tf.FixedLenFeature((), tf.string, default_value=''),
'image/format': tf.FixedLenFeature((), tf.string, default_value='png'),
'image/class/label': tf.FixedLenFeature(
[], tf.int64, default_value=tf.zeros([], dtype=tf.int64)),
}
items_to_handlers = {
'image': slim.tfexample_decoder.Image(),
'label': slim.tfexample_decoder.Tensor('image/class/label'),
}
decoder = slim.tfexample_decoder.TFExampleDecoder(
keys_to_features, items_to_handlers)
labels_to_names = None
if dataset_utils.has_labels(dataset_dir):
labels_to_names = dataset_utils.read_label_file(dataset_dir)
return slim.dataset.Dataset(
data_sources=file_pattern,
reader=reader,
decoder=decoder,
num_samples=SPLITS_TO_SIZES[split_name],
items_to_descriptions=_ITEMS_TO_DESCRIPTIONS,
num_classes=_NUM_CLASSES,
labels_to_names=labels_to_names)
def get_split(self, split_name):
file_pattern = self.dataset_name + '_' + split_name + '.tfrecord'
file_pattern = os.path.join(self.dataset_storage_location,
file_pattern)
print("file path.." + file_pattern)
reader = tf.TFRecordReader
keys_to_features = {
'image/encoded':
tf.FixedLenFeature((), tf.string, default_value=''),
'image/format':
tf.FixedLenFeature((), tf.string, default_value='png'),
'image/class/label':
tf.FixedLenFeature([],
tf.int64,
default_value=tf.zeros([], dtype=tf.int64)),
}
items_to_handlers = {
'image':
slim.tfexample_decoder.Image(
shape=[self.image_size, self.image_size, self.channels]),
'label':
slim.tfexample_decoder.Tensor('image/class/label'),
}
decoder = slim.tfexample_decoder.TFExampleDecoder(
keys_to_features, items_to_handlers)
labels_to_names = None
label_file_name = self.dataset_name + '_labels.txt'
if dataset_utils.has_labels(self.dataset_storage_location,
label_file_name):
labels_to_names = dataset_utils.read_label_file(
self.dataset_storage_location, label_file_name)
total_number_of_images = self.get_total_number_of_images(
split_name, self.dataset_name)
return slim.dataset.Dataset(
data_sources=file_pattern,
reader=reader,
decoder=decoder,
items_to_descriptions=self.dataset_description,
num_classes=self.number_of_classes,
num_samples=total_number_of_images,
labels_to_names=labels_to_names)
def get_split(split_name, dataset_dir, file_pattern=None, reader=None):
if split_name not in SPLITS_TO_SIZES:
raise ValueError('split name %s was not in recognized.' % split_name)
if not file_pattern:
file_pattern = _FILE_PATTERN
file_pattern = os.path.join(dataset_dir, file_pattern % split_name)
if reader is None:
reader = tf.TFRecordReader
key_to_features = {
'image/encoded':
tf.FixedLenFeature((), tf.string, default_value=''),
'image/label':
tf.FixedLenFeature([],
tf.int64,
default_value=tf.zeros([], dtype=tf.int64)),
'image/filename':
tf.FixedLenFeature((), tf.string, default_value=''),
'image/format':
tf.FixedLenFeature((), tf.string, default_value='jpg')
}
items_to_handlers = {
'image': slim.tfexample_decoder.Image('image/encoded'),
'label': slim.tfexample_decoder.Tensor('image/label'),
'filename': slim.tfexample_decoder.Tensor('image/filename')
}
decoder = slim.tfexample_decoder.TFExampleDecoder(key_to_features,
items_to_handlers)
labels_to_name = None
if dataset_utils.has_labels(dataset_dir):
labels_to_name = dataset_utils.read_label_file(dataset_dir)
return slim.dataset.Dataset(data_sources=file_pattern,
reader=reader,
decoder=decoder,
num_samples=SPLITS_TO_SIZES[split_name],
items_to_descriptions=_ITEMS_TO_DESCRIPTIONS,
num_classes=_NUM_CLASSES,
labels_to_names=labels_to_name)
def get_split(split_name, dataset_dir, file_pattern, reader, split_to_sizes,
items_to_descriptions, num_classes):
"""Gets a dataset tuple with instructions for reading Pascal VOC dataset.
Args:
split_name: A train/test split name.
dataset_dir: The base directory of the dataset sources.
file_pattern: The file pattern to use when matching the dataset sources.
It is assumed that the pattern contains a '%s' string so that the split
name can be inserted.
reader: The TensorFlow reader type.
Returns:
A `Dataset` namedtuple.
Raises:
ValueError: if `split_name` is not a valid train/test split.
"""
if split_name not in split_to_sizes:
raise ValueError('split name %s was not recognized.' % split_name)
print(split_name)
# file_pattern = os.path.join(dataset_dir, file_pattern % split_name)
file_pattern = os.path.join(dataset_dir, file_pattern)
print(file_pattern)
# Allowing None in the signature so that dataset_factory can use the default.
if reader is None:
reader = tf.TFRecordReader
# Features in Pascal VOC TFRecords.
# 'images/label': int64_feature(label),
# 'images/nc_roi': bytes_feature(nc_roi_data),
# 'images/art_roi': bytes_feature(art_roi_data),
# 'images/pv_roi': bytes_feature(pv_roi_data),
# 'images/nc_patch': bytes_feature(nc_patch_data),
# 'images/art_patch': bytes_feature(art_patch_data),
# 'images/pv_patch': bytes_feature(pv_patch_data),
# 'images/format': bytes_feature(image_format)}))
keys_to_features = {
'images/nc_roi': tf.FixedLenFeature((), tf.string, default_value=''),
'images/art_roi': tf.FixedLenFeature((), tf.string, default_value=''),
'images/pv_roi': tf.FixedLenFeature((), tf.string, default_value=''),
'images/nc_patch': tf.FixedLenFeature((), tf.string, default_value=''),
'images/art_patch': tf.FixedLenFeature((), tf.string,
default_value=''),
'images/pv_patch': tf.FixedLenFeature((), tf.string, default_value=''),
'images/format': tf.FixedLenFeature((),
tf.string,
default_value='jpeg'),
'images/label': tf.FixedLenFeature([1], tf.int64),
}
items_to_handlers = {
'nc_roi':
slim.tfexample_decoder.Image('images/nc_roi', 'images/format'),
'art_roi':
slim.tfexample_decoder.Image('images/art_roi', 'images/format'),
'pv_roi':
slim.tfexample_decoder.Image('images/pv_roi', 'images/format'),
'nc_patch':
slim.tfexample_decoder.Image('images/nc_patch', 'images/format'),
'art_patch':
slim.tfexample_decoder.Image('images/art_patch', 'images/format'),
'pv_patch':
slim.tfexample_decoder.Image('images/pv_patch', 'images/format'),
'label':
slim.tfexample_decoder.Tensor('images/label')
}
decoder = slim.tfexample_decoder.TFExampleDecoder(keys_to_features,
items_to_handlers)
labels_to_names = None
if dataset_utils.has_labels(dataset_dir):
labels_to_names = dataset_utils.read_label_file(dataset_dir)
# else:
# labels_to_names = create_readable_names_for_imagenet_labels()
# dataset_utils.write_label_file(labels_to_names, dataset_dir)
return slim.dataset.Dataset(data_sources=file_pattern,
reader=reader,
decoder=decoder,
num_samples=split_to_sizes[split_name],
items_to_descriptions=items_to_descriptions,
num_classes=num_classes,
labels_to_names=labels_to_names)