blob: 9160876cc3bce1d557dc47e1546203b8caecd555 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
|
"""Normalization layers for transfromers.
Copied from lucidrains:
https://github.com/lucidrains/x-transformers/blob/main/x_transformers/x_transformers.py
"""
from typing import Callable, Dict
import torch
from torch import nn
from torch import Tensor
class Rezero(nn.Module):
def __init__(self, fn: Callable) -> None:
super().__init__()
self.fn = fn
self.g = nn.Parameter(torch.zeros(1))
def forward(self, x: Tensor, **kwargs: Dict) -> Tensor:
x, *rest = self.fn(x, **kwargs)
return (x * self.g, *rest)
class ScaleNorm(nn.Module):
def __init__(self, dim: int, eps: float = 1.0e-5) -> None:
super().__init__()
self.scale = dim ** -0.5
self.eps = eps
self.g = nn.Parameter(torch.ones(1))
def forward(self, x: Tensor) -> Tensor:
norm = torch.norm(x, dim=-1, keepdim=True) * self.scale
return x / norm.clamp(min=self.eps) self.g
|
