유형 별칭

다음 유형 별칭은 전역적으로 사용할 수 있습니다.

  • 선언

    public typealias TensorFlowNumeric = TensorFlowScalar & Numeric
  • 선언

    public typealias TensorFlowSignedNumeric = TensorFlowScalar & SignedNumeric
  • 선언

    public typealias TensorFlowInteger = TensorFlowScalar & BinaryInteger
  • 선언

    public typealias ParameterInitializer<Scalar> = (TensorShape) -> Tensor<Scalar> where Scalar : TensorFlowScalar
  • 추론에 적합한 배치 모음으로, samples 에서 batchSize 배치로 샘플을 추출합니다.

    선언

    public typealias NonuniformInferenceBatches<Samples: Collection> = Slices<
      Sampling<Samples, [Samples.Index]>
    >
  • 선언

    public typealias OptimizerCallback = (inout OptimizerWeightStepState, inout OptimizerState) -> Void
  • 선언

    public typealias TensorFlowSeed = (graph: Int32, op: Int32)
  • 선언

    public typealias Raw = _Raw
  • 선언

    public typealias BasicRNN<Scalar> = RecurrentLayer<BasicRNNCell<Scalar>> where Scalar : TensorFlowFloatingPoint
  • 선언

    public typealias LSTM<Scalar> = RecurrentLayer<LSTMCell<Scalar>> where Scalar : TensorFlowFloatingPoint
  • 선언

    public typealias GRU<Scalar> = RecurrentLayer<GRUCell<Scalar>> where Scalar : TensorFlowFloatingPoint
  • 선언

    public typealias BidirectionalBasicRNN<Scalar> = BidirectionalRecurrentLayer<BasicRNNCell<Scalar>> where Scalar : TensorFlowFloatingPoint
  • 선언

    public typealias BidirectionalLSTM<Scalar> = BidirectionalRecurrentLayer<LSTMCell<Scalar>> where Scalar : TensorFlowFloatingPoint
  • 선언

    public typealias BidirectionalGRU<Scalar> = BidirectionalRecurrentLayer<GRUCell<Scalar>> where Scalar : TensorFlowFloatingPoint
  • 선언

    public typealias RNNCell = RecurrentLayerCell
  • RNN

    선언

    public typealias RNN = RecurrentLayer
  • 선언

    public typealias SimpleRNNCell = BasicRNNCell
  • 선언

    public typealias SimpleRNN = BasicRNN
  • 3개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential3<L1: Module, L2: Layer, L3: Layer> = Sequential<L1, Sequential<L2, L3>>
      where L1.Output == L2.Input, L2.Output == L3.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar
  • 4개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential4<L1: Module, L2: Layer, L3: Layer, L4: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, L4>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar
  • 5개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential5<L1: Module, L2: Layer, L3: Layer, L4: Layer, L5: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, Sequential<L4, L5>>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input, L4.Output == L5.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar, L4.TangentVector.VectorSpaceScalar == L5.TangentVector.VectorSpaceScalar
  • 6개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential6<L1: Module, L2: Layer, L3: Layer, L4: Layer, L5: Layer, L6: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, Sequential<L4, Sequential<L5, L6>>>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input, L4.Output == L5.Input, L5.Output == L6.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar, L4.TangentVector.VectorSpaceScalar == L5.TangentVector.VectorSpaceScalar, L5.TangentVector.VectorSpaceScalar == L6.TangentVector.VectorSpaceScalar
  • 7개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential7<L1: Module, L2: Layer, L3: Layer, L4: Layer, L5: Layer, L6: Layer, L7: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, Sequential<L4, Sequential<L5, Sequential<L6, L7>>>>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input, L4.Output == L5.Input, L5.Output == L6.Input, L6.Output == L7.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar, L4.TangentVector.VectorSpaceScalar == L5.TangentVector.VectorSpaceScalar, L5.TangentVector.VectorSpaceScalar == L6.TangentVector.VectorSpaceScalar, L6.TangentVector.VectorSpaceScalar == L7.TangentVector.VectorSpaceScalar
  • 8개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential8<L1: Module, L2: Layer, L3: Layer, L4: Layer, L5: Layer, L6: Layer, L7: Layer, L8: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, Sequential<L4, Sequential<L5, Sequential<L6, Sequential<L7, L8>>>>>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input, L4.Output == L5.Input, L5.Output == L6.Input, L6.Output == L7.Input, L7.Output == L8.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar, L4.TangentVector.VectorSpaceScalar == L5.TangentVector.VectorSpaceScalar, L5.TangentVector.VectorSpaceScalar == L6.TangentVector.VectorSpaceScalar, L6.TangentVector.VectorSpaceScalar == L7.TangentVector.VectorSpaceScalar, L7.TangentVector.VectorSpaceScalar == L8.TangentVector.VectorSpaceScalar
  • 9개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential9<L1: Module, L2: Layer, L3: Layer, L4: Layer, L5: Layer, L6: Layer, L7: Layer, L8: Layer, L9: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, Sequential<L4, Sequential<L5, Sequential<L6, Sequential<L7, Sequential<L8, L9>>>>>>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input, L4.Output == L5.Input, L5.Output == L6.Input, L6.Output == L7.Input, L7.Output == L8.Input, L8.Output == L9.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar, L4.TangentVector.VectorSpaceScalar == L5.TangentVector.VectorSpaceScalar, L5.TangentVector.VectorSpaceScalar == L6.TangentVector.VectorSpaceScalar, L6.TangentVector.VectorSpaceScalar == L7.TangentVector.VectorSpaceScalar, L7.TangentVector.VectorSpaceScalar == L8.TangentVector.VectorSpaceScalar, L8.TangentVector.VectorSpaceScalar == L9.TangentVector.VectorSpaceScalar
  • 10개의 레이어를 순차적으로 구성하는 레이어입니다.

    선언

    public typealias Sequential10<L1: Module, L2: Layer, L3: Layer, L4: Layer, L5: Layer, L6: Layer, L7: Layer, L8: Layer, L9: Layer, L10: Layer> = Sequential<L1, Sequential<L2, Sequential<L3, Sequential<L4, Sequential<L5, Sequential<L6, Sequential<L7, Sequential<L8, Sequential<L9, L10>>>>>>>>>
      where L1.Output == L2.Input, L2.Output == L3.Input, L3.Output == L4.Input, L4.Output == L5.Input, L5.Output == L6.Input, L6.Output == L7.Input, L7.Output == L8.Input, L8.Output == L9.Input, L9.Output == L10.Input,
            L1.TangentVector.VectorSpaceScalar == L2.TangentVector.VectorSpaceScalar, L2.TangentVector.VectorSpaceScalar == L3.TangentVector.VectorSpaceScalar, L3.TangentVector.VectorSpaceScalar == L4.TangentVector.VectorSpaceScalar, L4.TangentVector.VectorSpaceScalar == L5.TangentVector.VectorSpaceScalar, L5.TangentVector.VectorSpaceScalar == L6.TangentVector.VectorSpaceScalar, L6.TangentVector.VectorSpaceScalar == L7.TangentVector.VectorSpaceScalar, L7.TangentVector.VectorSpaceScalar == L8.TangentVector.VectorSpaceScalar, L8.TangentVector.VectorSpaceScalar == L9.TangentVector.VectorSpaceScalar, L9.TangentVector.VectorSpaceScalar == L10.TangentVector.VectorSpaceScalar