MCPcopy
hub / github.com/openai/improved-diffusion / normalization

Function normalization

improved_diffusion/nn.py:93–100  ·  view source on GitHub ↗

Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.

(channels)

Source from the content-addressed store, hash-verified

91
92
93def normalization(channels):
94 """
95 Make a standard normalization layer.
96
97 :param channels: number of input channels.
98 :return: an nn.Module for normalization.
99 """
100 return GroupNorm32(32, channels)
101
102
103def timestep_embedding(timesteps, dim, max_period=10000):

Callers 3

__init__Method · 0.85
__init__Method · 0.85
__init__Method · 0.85

Calls 1

GroupNorm32Class · 0.85

Tested by

no test coverage detected