`validation_split` argument, for returning both dataset splits at once,
        as a tuple.
    *   Added `tf.keras.utils.split_dataset` utility to split a `Dataset` object
        or a list/tuple of arrays into two `Dataset` objects (e.g. train/test).
    *   Added step granularity to `BackupAndRestore` callback for handling
        distributed training failures & restarts. The training state can now be

block;\n  width: 100%;\n  padding: 0;\n  overflow: hidden;\n}\n\n.embed-responsive::before {\n  display: block;\n  content: \"\";\n}\n\n.embed-responsive .embed-responsive-item,\n.embed-responsive iframe,\n.embed-responsive embed,\n.embed-responsive object,\n.embed-responsive video {\n  position: absolute;\n  top: 0;\n  bottom: 0;\n  left: 0;\n  width: 100%;\n  height: 100%;\n  border: 0;\n}\n\n.embed-responsive-21by9::before {\n  padding-top: 42.857143%;\n}\n\n.embed-responsive-16by9::before {\n  padding-top:...