Package org.tensorflow.op.nn

Class Conv2d.Options

java.lang.Object

org.tensorflow.op.nn.Conv2d.Options

Enclosing class:

Conv2d<T extends java.lang.Number>

public static class Conv2d.Options
extends java.lang.Object

Optional attributes for Conv2d

Method Summary

Modifier and Type	Method	Description
Conv2d.Options	dataFormat(java.lang.String dataFormat)
Conv2d.Options	dilations(java.util.List<java.lang.Long> dilations)
Conv2d.Options	useCudnnOnGpu(java.lang.Boolean useCudnnOnGpu)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

useCudnnOnGpu

public Conv2d.Options useCudnnOnGpu(java.lang.Boolean useCudnnOnGpu)

Parameters:

useCudnnOnGpu -

dataFormat

public Conv2d.Options dataFormat(java.lang.String dataFormat)

Parameters:

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, height, width, channels]. Alternatively, the format could be "NCHW", the data storage order of: [batch, channels, height, width].

dilations

public Conv2d.Options dilations(java.util.List<java.lang.Long> dilations)

Parameters:

dilations - 1-D tensor of length 4. The dilation factor for each dimension of `input`. If set to k > 1, there will be k-1 skipped cells between each filter element on that dimension. The dimension order is determined by the value of `data_format`, see above for details. Dilations in the batch and depth dimensions must be 1.