tf.contrib.checkpoint.Mapping

An append-only trackable mapping data structure with string keys.

tf.contrib.checkpoint.Mapping(
    *args, **kwargs
)

Maintains checkpoint dependencies on its contents (which must also be trackable), named based on its keys.

Note that once a key has been added, it may not be deleted or replaced. If names may not be unique, see tf.contrib.checkpoint.UniqueNameTracker.

Attributes
`layers`
`losses`	Aggregate losses from any `Layer` instances.
`non_trainable_variables`
`non_trainable_weights`
`trainable`
`trainable_variables`
`trainable_weights`
`updates`	Aggregate updates from any `Layer` instances.
`variables`
`weights`

Methods

get(
    key, default=None
)

D.get(k[,d]) -> D[k] if k in D, else d. d defaults to None.

items()

D.items() -> a set-like object providing a view on D's items

keys()

D.keys() -> a set-like object providing a view on D's keys

update(
    *args, **kwargs
)

values()

D.values() -> an object providing a view on D's values

__contains__(
    key
)

__eq__(
    other
)

Return self==value.

__getitem__(
    key
)

__iter__()

__len__()