QuantizedRelu

public final class QuantizedRelu

Computes Quantized Rectified Linear: `max(features, 0)`

Constants

String OP_NAME The name of this op, as known by TensorFlow core engine

Public Methods

Output<U>
activations()
Has the same output shape as "features".
static <U extends TType> QuantizedRelu<U>
create(Scope scope, Operand<? extends TType> features, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType)
Factory method to create a class wrapping a new QuantizedRelu operation.
Output<TFloat32>
maxActivations()
The float value that the highest quantized value represents.
Output<TFloat32>
minActivations()
The float value that the lowest quantized value represents.

Inherited Methods

Constants

public static final String OP_NAME

The name of this op, as known by TensorFlow core engine

Constant Value: "QuantizedRelu"

Public Methods

public Output<U> activations ()

Has the same output shape as "features".

public static QuantizedRelu<U> create (Scope scope, Operand<? extends TType> features, Operand<TFloat32> minFeatures, Operand<TFloat32> maxFeatures, Class<U> outType)

Factory method to create a class wrapping a new QuantizedRelu operation.

Parameters
scope current scope
minFeatures The float value that the lowest quantized value represents.
maxFeatures The float value that the highest quantized value represents.
Returns
  • a new instance of QuantizedRelu

public Output<TFloat32> maxActivations ()

The float value that the highest quantized value represents.

public Output<TFloat32> minActivations ()

The float value that the lowest quantized value represents.