MCPcopy
hub / github.com/openai/guided-diffusion / timestep_embedding

Function timestep_embedding

guided_diffusion/nn.py:103–121  ·  view source on GitHub ↗

Create sinusoidal timestep embeddings. :param timesteps: a 1-D Tensor of N indices, one per batch element. These may be fractional. :param dim: the dimension of the output. :param max_period: controls the minimum frequency of the embeddings. :return: an [N

(timesteps, dim, max_period=10000)

Source from the content-addressed store, hash-verified

101
102
103def timestep_embedding(timesteps, dim, max_period=10000):
104 """
105 Create sinusoidal timestep embeddings.
106
107 :param timesteps: a 1-D Tensor of N indices, one per batch element.
108 These may be fractional.
109 :param dim: the dimension of the output.
110 :param max_period: controls the minimum frequency of the embeddings.
111 :return: an [N x dim] Tensor of positional embeddings.
112 """
113 half = dim // 2
114 freqs = th.exp(
115 -math.log(max_period) * th.arange(start=0, end=half, dtype=th.float32) / half
116 ).to(device=timesteps.device)
117 args = timesteps[:, None].float() * freqs[None]
118 embedding = th.cat([th.cos(args), th.sin(args)], dim=-1)
119 if dim % 2:
120 embedding = th.cat([embedding, th.zeros_like(embedding[:, :1])], dim=-1)
121 return embedding
122
123
124def checkpoint(func, inputs, params, flag):

Callers 2

forwardMethod · 0.85
forwardMethod · 0.85

Calls 1

logMethod · 0.80

Tested by

no test coverage detected