tfr.keras.pipeline.NullDatasetBuilder

A no-op wrapper of datasets and signatures.

tfr.keras.pipeline.NullDatasetBuilder(
    train_dataset, valid_dataset, signatures=None
)

train_dataset = tf.data.Dataset(...)
valid_dataset = tf.data.Dataset(...)
dataset_builder = NullDatasetBuilder(train_dataset, valid_dataset)

Args
`train_dataset`	A `tf.data.Dataset` for training.
`valid_dataset`	A `tf.data.Dataset` for validation.
`signatures`	A dict of signatures that formulate the model in functions that render the input data with given types. When None, no signatures assigned.

Methods

build_signatures(
    *arg, **kwargs
) -> Any

See AbstractDatasetBuilder.

build_train_dataset(
    *arg, **kwargs
) -> tf.data.Dataset

See AbstractDatasetBuilder.

build_valid_dataset(
    *arg, **kwargs
) -> tf.data.Dataset

See AbstractDatasetBuilder.