diff options
Diffstat (limited to 'text_recognizer/networks/transformer')
-rw-r--r-- | text_recognizer/networks/transformer/norm.py | 16 |
1 files changed, 14 insertions, 2 deletions
diff --git a/text_recognizer/networks/transformer/norm.py b/text_recognizer/networks/transformer/norm.py index 537246d..4cd3b5b 100644 --- a/text_recognizer/networks/transformer/norm.py +++ b/text_recognizer/networks/transformer/norm.py @@ -4,7 +4,7 @@ Copied from lucidrains: https://github.com/lucidrains/x-transformers/blob/main/x_transformers/x_transformers.py """ -from typing import Dict, Type +from typing import Dict, Optional, Type import torch from torch import nn @@ -29,12 +29,24 @@ class RMSNorm(nn.Module): class PreNorm(nn.Module): """Applies layer normalization then function.""" - def __init__(self, normalized_shape: int, fn: Type[nn.Module]) -> None: + def __init__( + self, + normalized_shape: int, + fn: Type[nn.Module], + context_dim: Optional[int] = None, + ) -> None: super().__init__() self.norm = nn.LayerNorm(normalized_shape) self.fn = fn + self.norm_context = ( + nn.LayerNorm(context_dim) if context_dim is not None else None + ) def forward(self, x: Tensor, **kwargs) -> Tensor: """Applies pre norm.""" x = self.norm(x) + if self.norm_context is not None: + context = kwargs["context"] + normed_context = self.norm_context(context) + kwargs.update(context=normed_context) return self.fn(x, **kwargs) |