class documentation

Base class for fiftyone.utils.data.importers.LabeledImageDatasetImporter instances that import tf.train.Example protos containing labeled images.

This class assumes that the input TFRecords only contain the images themselves and not their paths on disk, and, therefore, the images are read in-memory and written to the provided images_dir during import.

Parameters
dataset_dirthe dataset directory. If omitted, tf_records_path must be provided
tf_records_path

an optional parameter that enables explicit control over the location of the TF records. Can be any of the following:

  • a filename like "tf.records" or glob pattern like "*.records-*-of-*" specifying the location of the records in dataset_dir
  • an absolute filepath or glob pattern for the records. In this case, dataset_dir has no effect on the location of the records

If None, the parameter will default to *record*

images_dirthe directory in which the images will be written. If not provided, the images will be unpacked into dataset_dir
image_formatthe image format to use to write the images to disk. By default, fiftyone.config.default_image_ext is used
force_rgbwhether to force convert all images to RGB
max_samplesa maximum number of samples to import. By default, all samples are imported
Method __init__ Undocumented
Method __iter__ Undocumented
Method __next__ Returns information about the next sample in the dataset.
Method close Performs any necessary actions after the last sample has been imported.
Method setup Performs any necessary setup before importing the first sample in the dataset.
Instance Variable force_rgb Undocumented
Instance Variable image_format Undocumented
Instance Variable images_dir Undocumented
Instance Variable tf_records_path Undocumented
Property has_dataset_info Whether this importer produces a dataset info dictionary.
Property has_image_metadata Whether this importer produces fiftyone.core.metadata.ImageMetadata instances for each image.
Method _make_sample_parser Returns a TFRecordSampleParser instance for parsing TFRecords read by this importer.
Instance Variable _dataset_ingestor Undocumented
Instance Variable _iter_dataset_ingestor Undocumented
Instance Variable _sample_parser Undocumented

Inherited from LabeledImageDatasetImporter:

Property label_cls The fiftyone.core.labels.Label class(es) returned by this importer.

Inherited from DatasetImporter (via LabeledImageDatasetImporter):

Method __enter__ Undocumented
Method __exit__ Undocumented
Method __len__ The total number of samples that will be imported.
Method get_dataset_info Returns the dataset info for the dataset.
Method _preprocess_list Internal utility that preprocesses the given list---which is presumed to be a list defining the samples that should be imported---by applying the values of the shuffle, seed, and max_samples parameters of the importer.

Inherited from ImportPathsMixin (via LabeledImageDatasetImporter, DatasetImporter):

Static Method _load_data_map Helper function that parses either a data directory or a data manifest file into a UUID -> filepath map.
Static Method _parse_data_path Helper function that computes default values for the data_path parameter supported by many importers.
Static Method _parse_labels_path Helper function that computes default values for the labels_path parameter supported by many importers.
def __init__(self, dataset_dir=None, tf_records_path=None, images_dir=None, image_format=None, force_rgb=False, max_samples=None): (source)
def __next__(self): (source)

Returns information about the next sample in the dataset.

Returns
an(image_path, image_metadata, label)
tuple
, where
Raises
StopIterationif there are no more samples to import
def close(self, *args): (source)

Performs any necessary actions after the last sample has been imported.

This method is called when the importer's context manager interface is exited, DatasetImporter.__exit__.

Parameters
*argsthe arguments to DatasetImporter.__exit__
def setup(self): (source)

Performs any necessary setup before importing the first sample in the dataset.

This method is called when the importer's context manager interface is entered, DatasetImporter.__enter__.

@property
has_dataset_info = (source)

Whether this importer produces a dataset info dictionary.

def _make_sample_parser(self): (source)

Returns a TFRecordSampleParser instance for parsing TFRecords read by this importer.

_dataset_ingestor = (source)

Undocumented

_iter_dataset_ingestor = (source)

Undocumented

_sample_parser = (source)

Undocumented