Class AvgPoolGrad.Inputs<T extends TNumber>

java.lang.Object
org.tensorflow.op.RawOpInputs<AvgPoolGrad<T>>
org.tensorflow.op.nn.AvgPoolGrad.Inputs<T>
Enclosing class:
AvgPoolGrad<T extends TNumber>

public static class AvgPoolGrad.Inputs<T extends TNumber> extends RawOpInputs<AvgPoolGrad<T>>
  • Field Details

    • origInputShape

      public final Operand<TInt32> origInputShape
      1-D. Shape of the original input to avg_pool.
    • grad

      public final Operand<T extends TNumber> grad
      4-D with shape [batch, height, width, channels]. Gradients w.r.t. the output of avg_pool.
    • ksize

      public final long[] ksize
      The size of the sliding window for each dimension of the input.
    • strides

      public final long[] strides
      The stride of the sliding window for each dimension of the input.
    • padding

      public final String padding
      The type of padding algorithm to use.
    • dataFormat

      public final String dataFormat
      Specify the data format of the input and output data. With the default format "NHWC", the data is stored in the order of: [batch, in_height, in_width, in_channels]. Alternatively, the format could be "NCHW", the data storage order of: [batch, in_channels, in_height, in_width].
    • T

      public final DataType T
      The T attribute
  • Constructor Details