TensorFlowLiteSwift Framework Reference

Options

public struct Options : Equatable, Hashable

Options for configuring the MetalDelegate.

  • Indicates whether the GPU delegate allows precision loss, such as allowing Float16 precision for a Float32 computation. The default is false.

    Declaration

    Swift

    public var isPrecisionLossAllowed: Bool

  • Undocumented

    Declaration

    Swift

    public var allowsPrecisionLoss: Bool { get set }

  • A type indicating how the current thread should wait for work on the GPU to complete. The default is passive.

    Declaration

    Swift

    public var waitType: ThreadWaitType

  • Indicates whether the GPU delegate allows execution of an 8-bit quantized model. The default is true.

    Declaration

    Swift

    public var isQuantizationEnabled: Bool

  • Creates a new instance with the default values.

    Declaration

    Swift

    public init()