# BNNS.RMSPropOptimizer An optimizer that uses the root mean square propagation (RMSProp) optimization method. ``` struct RMSPropOptimizer ``` ## Topics ### Creating an RMSProp Optimizer [`init(learningRate: Float, alpha: Float, epsilon: Float, centered: Bool, momentum: Float, gradientScale: Float, regularizationScale: Float, clipsGradientsTo: ClosedRange?, regularizationFunction: BNNSOptimizerRegularizationFunction)`](/documentation/Accelerate/BNNS/RMSPropOptimizer/init(learningRate:alpha:epsilon:centered:momentum:gradientScale:regularizationScale:clipsGradientsTo:regularizationFunction:)) Returns a new RMSProp optimizer object. [`init(learningRate: Float, alpha: Float, epsilon: Float, centered: Bool, momentum: Float, gradientScale: Float, regularizationScale: Float, gradientClipping: BNNS.GradientClipping, regularizationFunction: BNNSOptimizerRegularizationFunction)`](/documentation/Accelerate/BNNS/RMSPropOptimizer/init(learningRate:alpha:epsilon:centered:momentum:gradientScale:regularizationScale:gradientClipping:regularizationFunction:)) Returns a new RMSProp optimizer object with gradient clipped by value or clipped by norm. ### Inspecting the Properties of an RMSProp Optimizer [`var learningRate: Float`](/documentation/Accelerate/BNNS/RMSPropOptimizer/learningRate) A value that specifies the learning rate. [`var alpha: Float`](/documentation/Accelerate/BNNS/RMSPropOptimizer/alpha) A constant that specifies smoothing. [`var epsilon: Float`](/documentation/Accelerate/BNNS/RMSPropOptimizer/epsilon) A term that the optimizer adds to the denominator. [`var centered: Bool`](/documentation/Accelerate/BNNS/RMSPropOptimizer/centered) A Boolean value that specifies whether to use the centered variant. [`var momentum: Float`](/documentation/Accelerate/BNNS/RMSPropOptimizer/momentum) The rate of momentum decay. [`var gradientScale: Float`](/documentation/Accelerate/BNNS/RMSPropOptimizer/gradientScale) A value that specifies the gradient scaling factor. [`var regularizationScale: Float`](/documentation/Accelerate/BNNS/RMSPropOptimizer/regularizationScale) A value that specifies the regularization scaling factor. [`var gradientBounds: ClosedRange?`](/documentation/Accelerate/BNNS/RMSPropOptimizer/gradientBounds) The values for the minimum and maximum gradients. [`var gradientClipping: BNNS.GradientClipping`](/documentation/Accelerate/BNNS/RMSPropOptimizer/gradientClipping) The gradient clipping. [`enum GradientClipping`](/documentation/Accelerate/BNNS/GradientClipping) Constants that describe clipping functions. [`var regularizationFunction: BNNSOptimizerRegularizationFunction`](/documentation/Accelerate/BNNS/RMSPropOptimizer/regularizationFunction) The variable that specifies the regularization function. [`var accumulatorCountMultiplier: Int`](/documentation/Accelerate/BNNS/RMSPropOptimizer/accumulatorCountMultiplier) The number of accumulators required for each parameter. ### Applying an Optimization Step ## Relationships ### Conforms To [`BNNSOptimizer`](/documentation/Accelerate/BNNSOptimizer) --- 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)