Rate this Page
★★★★★

GroupNorm#

classtorch.ao.nn.quantized.GroupNorm(num_groups,num_channels,weight,bias,scale,zero_point,eps=1e-05,affine=True,device=None,dtype=None)[source]#

This is the quantized version ofGroupNorm.

Additional args:

  • scale - quantization scale of the output, type: double.

  • zero_point - quantization zero point of the output, type: long.