<!--
{
  "documentType" : "article",
  "framework" : "MLCompute",
  "identifier" : "/documentation/MLCompute/training-and-validation",
  "metadataVersion" : "0.1.0",
  "role" : "collectionGroup",
  "title" : "Training and Validation"
}
-->

# Training and Validation

Create, train, and validate a graph to produce acceptable prediction results.

## Topics

### Devices

[`class MLCDevice`](/documentation/MLCompute/MLCDevice)

An object that represents the CPU or one or more GPUs the framework uses to execute a neural network.

### Graphs

[`class MLCGraph`](/documentation/MLCompute/MLCGraph)

A graph of layers you use to build a training or inference graph.

[`class MLCTrainingGraph`](/documentation/MLCompute/MLCTrainingGraph)

A training graph that you create from one or more graph objects plus additional layers you add directly to the training graph.

[`class MLCInferenceGraph`](/documentation/MLCompute/MLCInferenceGraph)

An inference graph created from one or more MLCGraph instances plus additional layers added directly to the inference graph.


---

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)