QuantizedReluX
tensorflow C++ API
tensorflow::ops::QuantizedReluX
Computes Quantized Rectified Linear X:min(max(features, 0), max_value)
Summary
Arguments:
- scope: A Scope object
- min_features: The float value that the lowest quantized value represents.
- max_features: The float value that the highest quantized value represents.
Returns:
Outputactivations: Has the same output shape as "features".Outputmin_activations: The float value that the lowest quantized value represents.Outputmax_activations: The float value that the highest quantized value represents.
QuantizedReluX block
Source link : https://github.com/EXPNUNI/enuSpaceTensorflow/blob/master/enuSpaceTensorflow/tf_nn.cpp
Argument:
- Scope scope : A Scope object (A scope is generated automatically each page. A scope is not connected.)
- Input min_features: connect Input node.
- Input max_features: connect Input node.
- QuantizedReluX::Attrs attrs:input QuantizedReluX::Attrs in value. ex)out_type_= DT_QUINT8;
Return:
- Output activations: Output object of QuantizedReluX class object.
- Output min_activations: Output object of QuantizedReluX class object.
- Output max_activations: Output object of QuantizedReluX class object.
Result:
- std::vector(Tensor) result_activations : Returned object of executed result by calling session.
- std::vector(Tensor) result_min_activations : Returned object of executed result by calling session.
- std::vector(Tensor) result_max_activations : Returned object of executed result by calling session.