com.microsoft - QLinearSoftmax#

QLinearSoftmax - 1 (com.microsoft)#

Version

This version of the operator has been available since version 1 of domain com.microsoft.

Summary

QLinearSoftmax computes the normalized exponential values for the given input: Softmax(input, axis) = Exp(input) / ReduceSum(Exp(input), axis=axis, keepdims=1) The input does not need to explicitly be a 2D vector. The “axis” attribute indicates the dimension along which QLinearSoftmax will be performed for onnx v.13+. or the dimension coerced to NxD Matrix for onnx v.12-. The output tensor has the same shape.

Attributes

  • axis: apply softmax to elements for dimensions axis,or all dims along with axis according to op-version Default value is ?.

  • opset (required): opset version of corresponding SoftMax. Default value is ?.

Inputs

  • X (heterogeneous) - T: The input tensor

  • X_scale (heterogeneous) - tensor(float): Scale of quantized input ‘X’. It must be a scalar.

  • x_zero_point (optional, heterogeneous) - T: Zero point tensor for input ‘X’.It must be a scalar.

  • y_scale (heterogeneous) - tensor(float): Scale of quantized output ‘Y’. It must be a scalar.

  • y_zero_point (heterogeneous) - T: Zero point tensor for output ‘Y’. It must be a scalar.

Outputs

  • Y (heterogeneous) - T: Output data tensor from pooling across the input tensor. The output tensor has the same rank as the input.

Examples