# Module: tf.contrib.checkpoint

<br />

|---|
| [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png) View source on GitHub](https://github.com/tensorflow/tensorflow/blob/v1.15.0/tensorflow/contrib/checkpoint/__init__.py) |

Tools for working with object-based checkpoints.

Visualization and inspection:

#### Managing dependencies:

Trackable data structures:

#### Checkpoint management:

Saving and restoring Python state:

## Classes

[`class CheckpointManager`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/train/CheckpointManager): Deletes old checkpoints.

[`class Checkpointable`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/Checkpointable): Manages dependencies on other objects.

[`class CheckpointableBase`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/CheckpointableBase): Base class for `Trackable` objects without automatic dependencies.

[`class CheckpointableObjectGraph`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/CheckpointableObjectGraph): A ProtocolMessage

[`class List`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/List): An append-only sequence type which is trackable.

[`class Mapping`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/Mapping): An append-only trackable mapping data structure with string keys.

[`class NoDependency`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/NoDependency): Allows attribute assignment to `Trackable` objects with no dependency.

[`class NumpyState`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/NumpyState): A trackable object whose NumPy array attributes are saved/restored.

[`class PythonStateWrapper`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/train/experimental/PythonState): A mixin for putting Python state in an object-based checkpoint.

[`class UniqueNameTracker`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/UniqueNameTracker): Adds dependencies on trackable objects with name hints.

## Functions

[`capture_dependencies(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/capture_dependencies): Capture variables created within this scope as `Template` dependencies.

[`dot_graph_from_checkpoint(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/dot_graph_from_checkpoint): Visualizes an object-based checkpoint (from [`tf.train.Checkpoint`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/train/Checkpoint)).

[`list_objects(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/list_objects): Traverse the object graph and list all accessible objects.

[`object_metadata(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/object_metadata): Retrieves information about the objects in a checkpoint.

[`split_dependency(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/checkpoint/split_dependency): Creates multiple dependencies with a synchronized save/restore.