Class DFunctionBatchNorm_F64

java.lang.Object
deepboof.impl.forward.standard.BaseFunction<T>
deepboof.impl.backward.standard.BaseDFunction<deepboof.tensors.Tensor_F64>
deepboof.impl.backward.standard.BaseDBatchNorm_F64
deepboof.impl.backward.standard.DFunctionBatchNorm_F64
All Implemented Interfaces:
DBatchNorm<deepboof.tensors.Tensor_F64>, DFunctionBatchNorm<deepboof.tensors.Tensor_F64>, deepboof.DFunction<deepboof.tensors.Tensor_F64>, deepboof.forward.BatchNorm, deepboof.Function<deepboof.tensors.Tensor_F64>

public class DFunctionBatchNorm_F64
extends BaseDBatchNorm_F64
implements DFunctionBatchNorm<deepboof.tensors.Tensor_F64>
Implementation of DFunctionBatchNorm for Tensor_F64. Intermediate variables are cached in the forward pass.
  • Constructor Details

    • DFunctionBatchNorm_F64

      public DFunctionBatchNorm_F64​(boolean requiresGammaBeta)
  • Method Details

    • createShapeVariables

      protected int[] createShapeVariables​(int[] shapeInput)
      Description copied from class: BaseDBatchNorm_F64
      Create the shape for all the variables which are being normalized
      Specified by:
      createShapeVariables in class BaseDBatchNorm_F64
      Parameters:
      shapeInput - Shape of input tensor (without mini-batch)
      Returns:
      shape of variables tensor
    • _forward

      public void _forward​(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 output)
      Specified by:
      _forward in class deepboof.impl.forward.standard.BaseFunction<deepboof.tensors.Tensor_F64>
    • forwardsEvaluate

      public void forwardsEvaluate​(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 output)
    • _backwards

      protected void _backwards​(deepboof.tensors.Tensor_F64 input, deepboof.tensors.Tensor_F64 dout, deepboof.tensors.Tensor_F64 gradientInput, List<deepboof.tensors.Tensor_F64> gradientParameters)
      Specified by:
      _backwards in class BaseDFunction<deepboof.tensors.Tensor_F64>