summaryrefslogtreecommitdiff
path: root/src/text_recognizer/networks/vision_transformer.py
blob: f227954e927970714015d9490a64316f20a986f9 (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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
"""VisionTransformer module.

Splits each image into patches and feeds them to a transformer.

"""

from typing import Dict, Optional, Tuple, Type

from einops import rearrange, reduce
from einops.layers.torch import Rearrange
from loguru import logger
import torch
from torch import nn
from torch import Tensor

from text_recognizer.networks.transformer import PositionalEncoding, Transformer
from text_recognizer.networks.util import configure_backbone


class VisionTransformer(nn.Module):
    """Linear projection+Transfomer for image to sequence prediction, sort of based on the ideas from ViT."""

    def __init__(
        self,
        num_encoder_layers: int,
        num_decoder_layers: int,
        hidden_dim: int,
        vocab_size: int,
        num_heads: int,
        max_len: int,
        expansion_dim: int,
        dropout_rate: float,
        trg_pad_index: int,
        mlp_dim: Optional[int] = None,
        patch_size: Tuple[int, int] = (28, 28),
        stride: Tuple[int, int] = (1, 14),
        activation: str = "gelu",
        backbone: Optional[str] = None,
        backbone_args: Optional[Dict] = None,
    ) -> None:
        super().__init__()

        self.patch_size = patch_size
        self.stride = stride
        self.trg_pad_index = trg_pad_index
        self.slidning_window = self._configure_sliding_window()
        self.character_embedding = nn.Embedding(vocab_size, hidden_dim)
        self.position_encoding = PositionalEncoding(hidden_dim, dropout_rate, max_len)
        self.mlp_dim = mlp_dim

        self.use_backbone = False
        if backbone is None:
            self.linear_projection = nn.Linear(
                self.patch_size[0] * self.patch_size[1], hidden_dim
            )
        else:
            self.backbone = configure_backbone(backbone, backbone_args)
            if mlp_dim:
                self.mlp = nn.Linear(mlp_dim, hidden_dim)
            self.use_backbone = True

        self.transformer = Transformer(
            num_encoder_layers,
            num_decoder_layers,
            hidden_dim,
            num_heads,
            expansion_dim,
            dropout_rate,
            activation,
        )

        self.head = nn.Sequential(nn.Linear(hidden_dim, vocab_size),)

    def _configure_sliding_window(self) -> nn.Sequential:
        return nn.Sequential(
            nn.Unfold(kernel_size=self.patch_size, stride=self.stride),
            Rearrange(
                "b (c h w) t -> b t c h w",
                h=self.patch_size[0],
                w=self.patch_size[1],
                c=1,
            ),
        )

    def _create_trg_mask(self, trg: Tensor) -> Tensor:
        # Move this outside the transformer.
        trg_pad_mask = (trg != self.trg_pad_index)[:, None, None]
        trg_len = trg.shape[1]
        trg_sub_mask = torch.tril(
            torch.ones((trg_len, trg_len), device=trg.device)
        ).bool()
        trg_mask = trg_pad_mask & trg_sub_mask
        return trg_mask

    def encoder(self, src: Tensor) -> Tensor:
        """Forward pass with the encoder of the transformer."""
        return self.transformer.encoder(src)

    def decoder(self, trg: Tensor, memory: Tensor, trg_mask: Tensor) -> Tensor:
        """Forward pass with the decoder of the transformer + classification head."""
        return self.head(
            self.transformer.decoder(trg=trg, memory=memory, trg_mask=trg_mask)
        )

    def _backbone(self, x: Tensor) -> Tensor:
        b, t = x.shape[:2]
        if self.use_backbone:
            x = rearrange(x, "b t c h w -> (b t) c h w", b=b, t=t)
            x = self.backbone(x)
            if self.mlp_dim:
                x = rearrange(x, "(b t) c h w -> b t (c h w)", b=b, t=t)
                x = self.mlp(x)
            else:
                x = rearrange(x, "(b t) h -> b t h", b=b, t=t)
        else:
            x = rearrange(x, "b t c h w -> b t (c h w)", b=b, t=t)
            x = self.linear_projection(x)
        return x

    def preprocess_input(self, src: Tensor) -> Tensor:
        """Encodes src with a backbone network and a positional encoding.

        Args:
            src (Tensor): Input tensor.

        Returns:
            Tensor: A input src to the transformer.

        """
        # If batch dimenstion is missing, it needs to be added.
        if len(src.shape) < 4:
            src = src[(None,) * (4 - len(src.shape))]
        src = self.slidning_window(src)  # .squeeze(-2)
        src = self._backbone(src)
        src = self.position_encoding(src)
        return src

    def preprocess_target(self, trg: Tensor) -> Tuple[Tensor, Tensor]:
        """Encodes target tensor with embedding and postion.

        Args:
            trg (Tensor): Target tensor.

        Returns:
            Tuple[Tensor, Tensor]: Encoded target tensor and target mask.

        """
        trg_mask = self._create_trg_mask(trg)
        trg = self.character_embedding(trg.long())
        trg = self.position_encoding(trg)
        return trg, trg_mask

    def forward(self, x: Tensor, trg: Tensor) -> Tensor:
        """Forward pass with vision transfomer."""
        src = self.preprocess_input(x)
        trg, trg_mask = self.preprocess_target(trg)
        out = self.transformer(src, trg, trg_mask=trg_mask)
        logits = self.head(out)
        return logits