Link Search Menu Expand Document

QuantizedMul


tensorflow C++ API

tensorflow::ops::QuantizedMul

Returns x * y element-wise, working on quantized buffers.


Summary

Arguments:

  • scope: A Scope object
  • min_x: The float value that the lowest quantizedxvalue represents.
  • max_x: The float value that the highest quantizedxvalue represents.
  • min_y: The float value that the lowest quantizedyvalue represents.
  • max_y: The float value that the highest quantizedyvalue represents.

Output

  • output z
  • output min_z: The float value that the lowest quantized output value represents.
  • output : max_z: The float value that the highest quantized output value represents.

NOTEQuantizedMulsupports limited forms of broadcasting. More about broadcasting here

Constructor

  • QuantizedMul(const ::tensorflow::Scope & scope, ::tensorflow::Input x, ::tensorflow::Input y, ::tensorflow::Input min_x, ::tensorflow::Input max_x, ::tensorflow::Input min_y, ::tensorflow::Input max_y, const QuantizedMul::Attrs & attrs)

Public attributes

  • tensorflow::Output z.
  • tensorflow::Output min_z.
  • tensorflow::Output max_z.

QuantizedMul block

Source link : https://github.com/EXPNUNI/enuSpaceTensorflow/blob/master/enuSpaceTensorflow/tf_math.cpp

Argument:

  • Scope scope : A Scope object (A scope is generated automatically each page. A scope is not connected.).
  • Input x : connect Input node.
  • Input y : connect Input node.
  • Input min_x : connect Input node.
  • Input max_x : connect Input node.
  • Input min_y : connect Input node.
  • Input max_y : connect Input node.
  • DataType out_type:Input DataType. ex)DT_QINT8

Return:

  • Output output: Output object of QuantizedMul class object.

Result:

  • std::vector(Tensor) product_result : Returned object of executed result by calling session.

Using Method

tensorflow -> error: No OpKernel was registered to support Op ‘QuantizedMul’ with these attrs.

Registered devices: [CPU,GPU], Registered kernels:

<no registered kernels>