<!--
{
  "availability" : [
    "iOS: 15.0.0 -",
    "iPadOS: 15.0.0 -",
    "macCatalyst: 15.0.0 -",
    "macOS: 11.0.0 -",
    "tvOS: 16.0.0 -",
    "visionOS: 1.0.0 -"
  ],
  "documentType" : "symbol",
  "framework" : "CreateML",
  "identifier" : "/documentation/CreateML/MLCheckpoint",
  "metadataVersion" : "0.1.0",
  "role" : "Structure",
  "symbol" : {
    "kind" : "Structure",
    "modules" : [
      "Create ML"
    ],
    "preciseIdentifier" : "s:8CreateML12MLCheckpointV"
  },
  "title" : "MLCheckpoint"
}
-->

# MLCheckpoint

The state of a model’s asynchronous training session at a specific point in
time during the feature extraction or training phase.

```
struct MLCheckpoint
```

## Topics

### Inspecting a checkpoint

[`phase`](/documentation/CreateML/MLCheckpoint/phase)

The training session’s phase when it created the checkpoint.

[`iteration`](/documentation/CreateML/MLCheckpoint/iteration)

The iteration number of a training session’s phase when it created the
checkpoint.

[`date`](/documentation/CreateML/MLCheckpoint/date)

The time when the training session created the checkpoint.

[`url`](/documentation/CreateML/MLCheckpoint/url)

The location of the checkpoint in the file system.

### Assessing a checkpoint

[`metrics`](/documentation/CreateML/MLCheckpoint/metrics)

Measurements of the model’s performance at the time the session saved
the checkpoint.

[`MLProgress.Metric`](/documentation/CreateML/MLProgress/Metric)

Metrics you use to evaluate a model’s performance during a training
session.

### Encoding and decoding a checkpoint

[`encode(to:)`](/documentation/CreateML/MLCheckpoint/encode(to:))

Encodes the checkpoint into the encoder.

[`init(from:)`](/documentation/CreateML/MLCheckpoint/init(from:))

Creates a new checkpoint by decoding from the decoder.


---

Copyright &copy; 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)