com.microsoft - GroupNorm#

GroupNorm - 1#

Version

  • name: GroupNorm (GitHub)

  • domain: com.microsoft

  • since_version: 1

  • function:

  • support_level: SupportType.COMMON

  • shape inference: True

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

Summary

Attributes

  • activation - INT (required) : Activation after group normalization: 0 for None, 1 for Swish

  • epsilon - FLOAT : The epsilon value to use to avoid division by zero

  • groups - INT (required) : The number of groups of channels. It should be a divisor of the number of channels C

Inputs

  • X (heterogeneous) - T:

  • gamma (heterogeneous) - M:

  • beta (heterogeneous) - M:

Outputs

  • Y (heterogeneous) - T:

Type Constraints

  • T in ( tensor(float), tensor(float16) ): Constrain input X and output Y types to float tensors.

  • M in ( tensor(float) ): Constrain gamma and beta to float tensors.

Examples