# BNNSOptimizerFunction A structure that contains optimizer functions. ``` struct BNNSOptimizerFunction ``` ## Topics ### Adam Optimizer Functions [`BNNSOptimizerFunctionAdam`](/documentation/Accelerate/BNNSOptimizerFunctionAdam) An optimizer function that updates parameters according to the Adam algorithm. [`BNNSOptimizerFunctionAdamWithClipping`](/documentation/Accelerate/BNNSOptimizerFunctionAdamWithClipping) An optimizer function that updates parameters according to the Adam algorithm and optionally clips the gradient by value or by norm. [`BNNSOptimizerFunctionAdamAMSGrad`](/documentation/Accelerate/BNNSOptimizerFunctionAdamAMSGrad) An optimizer function that updates parameters according to the AMSGrad variant of the Adam algorithm. [`BNNSOptimizerFunctionAdamAMSGradWithClipping`](/documentation/Accelerate/BNNSOptimizerFunctionAdamAMSGradWithClipping) An optimizer function that updates parameters according to the AMSGrad variant of the Adam algorithm and optionally clips the gradient by value or by norm. ### AdamW Optimizer Functions [`BNNSOptimizerFunctionAdamW`](/documentation/Accelerate/BNNSOptimizerFunctionAdamW) An optimizer function that updates parameters according to the AdamW algorithm. [`BNNSOptimizerFunctionAdamWWithClipping`](/documentation/Accelerate/BNNSOptimizerFunctionAdamWWithClipping) An optimizer function that updates parameters according to the AdamW algorithm and optionally clips the gradient by value or by norm. [`BNNSOptimizerFunctionAdamWAMSGrad`](/documentation/Accelerate/BNNSOptimizerFunctionAdamWAMSGrad) An optimizer function that updates parameters according to the AMSGrad variant of the AdamW algorithm. [`BNNSOptimizerFunctionAdamWAMSGradWithClipping`](/documentation/Accelerate/BNNSOptimizerFunctionAdamWAMSGradWithClipping) An optimizer function that updates parameters according to the AMSGrad variant of the AdamW algorithm and optionally clips the gradient by value or by norm. ### RMSProp Optimizer Functions [`BNNSOptimizerFunctionRMSProp`](/documentation/Accelerate/BNNSOptimizerFunctionRMSProp) An optimizer function that updates parameters according to the root mean square propagation (RMSProp) algorithm. [`BNNSOptimizerFunctionRMSPropWithClipping`](/documentation/Accelerate/BNNSOptimizerFunctionRMSPropWithClipping) An optimizer function that updates parameters according to the root mean square propagation (RMSProp) algorithm and optionally clips the gradient by value or by norm. ### SGD Optimizer Functions [`BNNSOptimizerFunctionSGDMomentum`](/documentation/Accelerate/BNNSOptimizerFunctionSGDMomentum) An optimizer function that updates parameters according to the stochastic gradient descent (SGD) with momentum algorithm. [`BNNSOptimizerFunctionSGDMomentumWithClipping`](/documentation/Accelerate/BNNSOptimizerFunctionSGDMomentumWithClipping) An optimizer function that updates parameters according to the stochastic gradient descent (SGD) with momentum algorithm and optionally clips the gradient by value or by norm. ### Raw Values [`rawValue`](/documentation/Accelerate/BNNSOptimizerFunction/rawValue) [`init(_:)`](/documentation/Accelerate/BNNSOptimizerFunction/init(_:)) [`init(rawValue:)`](/documentation/Accelerate/BNNSOptimizerFunction/init(rawValue:)) --- Copyright © 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)