<!--
{
  "availability" : [
    "iOS: 14.0.0 - 18.0.0",
    "iPadOS: 14.0.0 - 18.0.0",
    "macCatalyst: 14.0.0 - 18.0.0",
    "macOS: 11.0.0 - 15.0.0",
    "tvOS: 14.0.0 - 18.0.0",
    "visionOS: 1.0.0 - 2.0.0",
    "watchOS: 7.0.0 - 11.0.0"
  ],
  "documentType" : "symbol",
  "framework" : "Accelerate",
  "identifier" : "/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields",
  "metadataVersion" : "0.1.0",
  "role" : "Structure",
  "symbol" : {
    "kind" : "Structure",
    "modules" : [
      "Accelerate"
    ],
    "preciseIdentifier" : "c:@SA@BNNSOptimizerAdamWithClippingFields"
  },
  "title" : "BNNSOptimizerAdamWithClippingFields"
}
-->

# BNNSOptimizerAdamWithClippingFields

A structure that contains the fields of an Adam or AdamW optimizer that optionally clips the gradient by value or by norm.

```
struct BNNSOptimizerAdamWithClippingFields
```

## Topics

### Initializers

[`init(learning_rate:beta1:beta2:time_step:epsilon:gradient_scale:regularization_scale:regularization_func:clipping_func:clip_gradients_min:clip_gradients_max:clip_gradients_max_norm:clip_gradients_use_norm:)`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/init(learning_rate:beta1:beta2:time_step:epsilon:gradient_scale:regularization_scale:regularization_func:clipping_func:clip_gradients_min:clip_gradients_max:clip_gradients_max_norm:clip_gradients_use_norm:))

Returns a new Adam or AdamW optimizer fields structure from the specified parameters.

[`init()`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/init())

Returns a new Adam or AdamW optimizer fields structure.

### Instance Properties

[`learning_rate`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/learning_rate)

A value that specifies the learning rate.

[`beta1`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/beta1)

A value that specifies the first moment constant in the range 0 to 1.

[`beta2`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/beta2)

A value that specifies the second moment constant in the range 0 to 1.

[`time_step`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/time_step)

A value that’s at least 1 and represents the optimizer’s current time.

[`epsilon`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/epsilon)

An addition for the division in the parameter update stage.

[`gradient_scale`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/gradient_scale)

A value that specifies the gradient scaling factor.

[`regularization_scale`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/regularization_scale)

A value that specifies the regularization scaling factor.

[`regularization_func`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/regularization_func)

The variable that specifies the regularization function.

[`clipping_func`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/clipping_func)

The clipping function.

[`BNNSOptimizerClippingFunction`](/documentation/Accelerate/BNNSOptimizerClippingFunction)

Constants that describe clipping functions.

[`clip_gradients_min`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/clip_gradients_min)

The minimum clipping value for clipping by value.

[`clip_gradients_max`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/clip_gradients_max)

The maximum clipping value for clipping by value.

[`clip_gradients_max_norm`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/clip_gradients_max_norm)

The maximum Euclidean norm for clipping by norm and clipping by global norm.

[`clip_gradients_use_norm`](/documentation/Accelerate/BNNSOptimizerAdamWithClippingFields/clip_gradients_use_norm)

An optional value for a known Euclidean norm for clipping by global norm.


---

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)