com.microsoft - QLinearReduceMean#
QLinearReduceMean - 1 (com.microsoft)#
Version
domain: com.microsoft
since_version: 1
function:
support_level:
shape inference:
This version of the operator has been available since version 1 of domain com.microsoft.
Summary
Computes the mean of the low-precision input tensor’s element along the provided axes. The resulting tensor has the same rank as the input if keepdims equal 1. If keepdims equal 0, then the resulting tensor have the reduced dimension pruned. The above behavior is similar to numpy, with the exception that numpy default keepdims to False instead of True. Input and Output scales and zero points are used to requantize the output in a new range. This helps to improve accuracy as after ReduceMean operation the range of the output is expected to decrease.
"Output = Dequantize(Input) -> ReduceMean on fp32 data -> Quantize(output)",
Attributes
axes (required): A list of integers, along which to reduce. The default is to reduce over all the dimensions of the input tensor. Default value is
?
.keepdims (required): Keep the reduced dimension or not, default 1 mean keep reduced dimension. Default value is
?
.
Inputs
Between 4 and 5 inputs.
data (heterogeneous) - T: An input tensor.
data_scale (heterogeneous) - tensor(float): Input scale. It’s a scalar, which means a per-tensor/layer quantization.
data_zero_point (optional, heterogeneous) - T: Input zero point. Default value is 0 if it’s not specified. It’s a scalar, which means a per-tensor/layer quantization.
reduced_scale (heterogeneous) - tensor(float): Output scale. It’s a scalar, which means a per-tensor/layer quantization.
reduced_zero_point (optional, heterogeneous) - T: Output zero point. Default value is 0 if it’s not specified. It’s a scalar, which means a per-tensor/layer quantization.
Outputs
reduced (heterogeneous) - T: Reduced output tensor.
Examples