File size: 14,240 Bytes
bfa5175
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
'''
 * Copyright (c) 2022, salesforce.com, inc.
 * All rights reserved.
 * SPDX-License-Identifier: BSD-3-Clause
 * For full license text, see LICENSE.txt file in the repo root or https://opensource.org/licenses/BSD-3-Clause
 * By Junnan Li
 * Based on timm code base
 * https://github.com/rwightman/pytorch-image-models/tree/master/timm
'''

import torch
import torch.nn as nn
import torch.nn.functional as F
from functools import partial

from timm.models.vision_transformer import _cfg, PatchEmbed
from timm.models.registry import register_model
from timm.models.layers import trunc_normal_, DropPath
from timm.models.helpers import named_apply, adapt_input_conv

from fairscale.nn.checkpoint.checkpoint_activations import checkpoint_wrapper

class Mlp(nn.Module):
    """ MLP as used in Vision Transformer, MLP-Mixer and related networks
    """
    def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.):
        super().__init__()
        out_features = out_features or in_features
        hidden_features = hidden_features or in_features
        self.fc1 = nn.Linear(in_features, hidden_features)
        self.act = act_layer()
        self.fc2 = nn.Linear(hidden_features, out_features)
        self.drop = nn.Dropout(drop)

    def forward(self, x):
        x = self.fc1(x)
        x = self.act(x)
        x = self.drop(x)
        x = self.fc2(x)
        x = self.drop(x)
        return x


class Attention(nn.Module):
    def __init__(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0.):
        super().__init__()
        self.num_heads = num_heads
        head_dim = dim // num_heads
        # NOTE scale factor was wrong in my original version, can set manually to be compat with prev weights
        self.scale = qk_scale or head_dim ** -0.5
        self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
        self.attn_drop = nn.Dropout(attn_drop)
        self.proj = nn.Linear(dim, dim)
        self.proj_drop = nn.Dropout(proj_drop)
        self.attn_gradients = None
        self.attention_map = None
        
    def save_attn_gradients(self, attn_gradients):
        self.attn_gradients = attn_gradients
        
    def get_attn_gradients(self):
        return self.attn_gradients
    
    def save_attention_map(self, attention_map):
        self.attention_map = attention_map
        
    def get_attention_map(self):
        return self.attention_map
    
    def forward(self, x, register_hook=False):
        B, N, C = x.shape
        qkv = self.qkv(x).reshape(B, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)
        q, k, v = qkv[0], qkv[1], qkv[2]   # make torchscript happy (cannot use tensor as tuple)

        attn = (q @ k.transpose(-2, -1)) * self.scale
        attn = attn.softmax(dim=-1)
        attn = self.attn_drop(attn)
                
        if register_hook:
            self.save_attention_map(attn)
            attn.register_hook(self.save_attn_gradients)        

        x = (attn @ v).transpose(1, 2).reshape(B, N, C)
        x = self.proj(x)
        x = self.proj_drop(x)
        return x


class Block(nn.Module):

    def __init__(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0.,
                 drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm, use_grad_checkpointing=False):
        super().__init__()
        self.norm1 = norm_layer(dim)
        self.attn = Attention(
            dim, num_heads=num_heads, qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop)
        # NOTE: drop path for stochastic depth, we shall see if this is better than dropout here
        self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
        self.norm2 = norm_layer(dim)
        mlp_hidden_dim = int(dim * mlp_ratio)
        self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)

        if use_grad_checkpointing:
            self.attn = checkpoint_wrapper(self.attn)
            self.mlp = checkpoint_wrapper(self.mlp)

    def forward(self, x, register_hook=False):
        x = x + self.drop_path(self.attn(self.norm1(x), register_hook=register_hook))
        x = x + self.drop_path(self.mlp(self.norm2(x)))
        return x

    
class VisionTransformer(nn.Module):
    """ Vision Transformer
    A PyTorch impl of : `An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale`  -
        https://arxiv.org/abs/2010.11929
    """
    def __init__(self, img_size=224, patch_size=16, in_chans=3, num_classes=1000, embed_dim=768, depth=12,
                 num_heads=12, mlp_ratio=4., qkv_bias=True, qk_scale=None, representation_size=None,
                 drop_rate=0., attn_drop_rate=0., drop_path_rate=0., norm_layer=None, 
                 use_grad_checkpointing=False, ckpt_layer=0):
        """
        Args:
            img_size (int, tuple): input image size
            patch_size (int, tuple): patch size
            in_chans (int): number of input channels
            num_classes (int): number of classes for classification head
            embed_dim (int): embedding dimension
            depth (int): depth of transformer
            num_heads (int): number of attention heads
            mlp_ratio (int): ratio of mlp hidden dim to embedding dim
            qkv_bias (bool): enable bias for qkv if True
            qk_scale (float): override default qk scale of head_dim ** -0.5 if set
            representation_size (Optional[int]): enable and set representation layer (pre-logits) to this value if set
            drop_rate (float): dropout rate
            attn_drop_rate (float): attention dropout rate
            drop_path_rate (float): stochastic depth rate
            norm_layer: (nn.Module): normalization layer
        """
        super().__init__()
        self.num_features = self.embed_dim = embed_dim  # num_features for consistency with other models
        norm_layer = norm_layer or partial(nn.LayerNorm, eps=1e-6)

        self.patch_embed = PatchEmbed(
            img_size=img_size, patch_size=patch_size, in_chans=in_chans, embed_dim=embed_dim)

        num_patches = self.patch_embed.num_patches

        self.cls_token = nn.Parameter(torch.zeros(1, 1, embed_dim))
        self.pos_embed = nn.Parameter(torch.zeros(1, num_patches + 1, embed_dim))
        self.pos_drop = nn.Dropout(p=drop_rate)

        dpr = [x.item() for x in torch.linspace(0, drop_path_rate, depth)]  # stochastic depth decay rule
        self.blocks = nn.ModuleList([
            Block(
                dim=embed_dim, num_heads=num_heads, mlp_ratio=mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale,
                drop=drop_rate, attn_drop=attn_drop_rate, drop_path=dpr[i], norm_layer=norm_layer,
                use_grad_checkpointing=(use_grad_checkpointing and i>=depth-ckpt_layer)
            )
            for i in range(depth)])
        self.norm = norm_layer(embed_dim)

        trunc_normal_(self.pos_embed, std=.02)
        trunc_normal_(self.cls_token, std=.02)
        self.apply(self._init_weights)

    def _init_weights(self, m):
        if isinstance(m, nn.Linear):
            trunc_normal_(m.weight, std=.02)
            if isinstance(m, nn.Linear) and m.bias is not None:
                nn.init.constant_(m.bias, 0)
        elif isinstance(m, nn.LayerNorm):
            nn.init.constant_(m.bias, 0)
            nn.init.constant_(m.weight, 1.0)

    @torch.jit.ignore
    def no_weight_decay(self):
        return {'pos_embed', 'cls_token'}

    def forward(self, x, register_blk=-1):
        B = x.shape[0]
        x = self.patch_embed(x)

        cls_tokens = self.cls_token.expand(B, -1, -1)  # stole cls_tokens impl from Phil Wang, thanks
        x = torch.cat((cls_tokens, x), dim=1)
  
        x = x + self.pos_embed[:,:x.size(1),:]
        x = self.pos_drop(x)

        for i,blk in enumerate(self.blocks):
            x = blk(x, register_blk==i)
        x = self.norm(x)
        
        return x

    @torch.jit.ignore()
    def load_pretrained(self, checkpoint_path, prefix=''):
        _load_weights(self, checkpoint_path, prefix)
        

@torch.no_grad()
def _load_weights(model: VisionTransformer, checkpoint_path: str, prefix: str = ''):
    """ Load weights from .npz checkpoints for official Google Brain Flax implementation
    """
    import numpy as np

    def _n2p(w, t=True):
        if w.ndim == 4 and w.shape[0] == w.shape[1] == w.shape[2] == 1:
            w = w.flatten()
        if t:
            if w.ndim == 4:
                w = w.transpose([3, 2, 0, 1])
            elif w.ndim == 3:
                w = w.transpose([2, 0, 1])
            elif w.ndim == 2:
                w = w.transpose([1, 0])
        return torch.from_numpy(w)

    w = np.load(checkpoint_path)
    if not prefix and 'opt/target/embedding/kernel' in w:
        prefix = 'opt/target/'

    if hasattr(model.patch_embed, 'backbone'):
        # hybrid
        backbone = model.patch_embed.backbone
        stem_only = not hasattr(backbone, 'stem')
        stem = backbone if stem_only else backbone.stem
        stem.conv.weight.copy_(adapt_input_conv(stem.conv.weight.shape[1], _n2p(w[f'{prefix}conv_root/kernel'])))
        stem.norm.weight.copy_(_n2p(w[f'{prefix}gn_root/scale']))
        stem.norm.bias.copy_(_n2p(w[f'{prefix}gn_root/bias']))
        if not stem_only:
            for i, stage in enumerate(backbone.stages):
                for j, block in enumerate(stage.blocks):
                    bp = f'{prefix}block{i + 1}/unit{j + 1}/'
                    for r in range(3):
                        getattr(block, f'conv{r + 1}').weight.copy_(_n2p(w[f'{bp}conv{r + 1}/kernel']))
                        getattr(block, f'norm{r + 1}').weight.copy_(_n2p(w[f'{bp}gn{r + 1}/scale']))
                        getattr(block, f'norm{r + 1}').bias.copy_(_n2p(w[f'{bp}gn{r + 1}/bias']))
                    if block.downsample is not None:
                        block.downsample.conv.weight.copy_(_n2p(w[f'{bp}conv_proj/kernel']))
                        block.downsample.norm.weight.copy_(_n2p(w[f'{bp}gn_proj/scale']))
                        block.downsample.norm.bias.copy_(_n2p(w[f'{bp}gn_proj/bias']))
        embed_conv_w = _n2p(w[f'{prefix}embedding/kernel'])
    else:
        embed_conv_w = adapt_input_conv(
            model.patch_embed.proj.weight.shape[1], _n2p(w[f'{prefix}embedding/kernel']))
    model.patch_embed.proj.weight.copy_(embed_conv_w)
    model.patch_embed.proj.bias.copy_(_n2p(w[f'{prefix}embedding/bias']))
    model.cls_token.copy_(_n2p(w[f'{prefix}cls'], t=False))
    pos_embed_w = _n2p(w[f'{prefix}Transformer/posembed_input/pos_embedding'], t=False)
    if pos_embed_w.shape != model.pos_embed.shape:
        pos_embed_w = resize_pos_embed(  # resize pos embedding when different size from pretrained weights
            pos_embed_w, model.pos_embed, getattr(model, 'num_tokens', 1), model.patch_embed.grid_size)
    model.pos_embed.copy_(pos_embed_w)
    model.norm.weight.copy_(_n2p(w[f'{prefix}Transformer/encoder_norm/scale']))
    model.norm.bias.copy_(_n2p(w[f'{prefix}Transformer/encoder_norm/bias']))
#     if isinstance(model.head, nn.Linear) and model.head.bias.shape[0] == w[f'{prefix}head/bias'].shape[-1]:
#         model.head.weight.copy_(_n2p(w[f'{prefix}head/kernel']))
#         model.head.bias.copy_(_n2p(w[f'{prefix}head/bias']))
#     if isinstance(getattr(model.pre_logits, 'fc', None), nn.Linear) and f'{prefix}pre_logits/bias' in w:
#         model.pre_logits.fc.weight.copy_(_n2p(w[f'{prefix}pre_logits/kernel']))
#         model.pre_logits.fc.bias.copy_(_n2p(w[f'{prefix}pre_logits/bias']))
    for i, block in enumerate(model.blocks.children()):
        block_prefix = f'{prefix}Transformer/encoderblock_{i}/'
        mha_prefix = block_prefix + 'MultiHeadDotProductAttention_1/'
        block.norm1.weight.copy_(_n2p(w[f'{block_prefix}LayerNorm_0/scale']))
        block.norm1.bias.copy_(_n2p(w[f'{block_prefix}LayerNorm_0/bias']))
        block.attn.qkv.weight.copy_(torch.cat([
            _n2p(w[f'{mha_prefix}{n}/kernel'], t=False).flatten(1).T for n in ('query', 'key', 'value')]))
        block.attn.qkv.bias.copy_(torch.cat([
            _n2p(w[f'{mha_prefix}{n}/bias'], t=False).reshape(-1) for n in ('query', 'key', 'value')]))
        block.attn.proj.weight.copy_(_n2p(w[f'{mha_prefix}out/kernel']).flatten(1))
        block.attn.proj.bias.copy_(_n2p(w[f'{mha_prefix}out/bias']))
        for r in range(2):
            getattr(block.mlp, f'fc{r + 1}').weight.copy_(_n2p(w[f'{block_prefix}MlpBlock_3/Dense_{r}/kernel']))
            getattr(block.mlp, f'fc{r + 1}').bias.copy_(_n2p(w[f'{block_prefix}MlpBlock_3/Dense_{r}/bias']))
        block.norm2.weight.copy_(_n2p(w[f'{block_prefix}LayerNorm_2/scale']))
        block.norm2.bias.copy_(_n2p(w[f'{block_prefix}LayerNorm_2/bias']))

            
def interpolate_pos_embed(pos_embed_checkpoint, visual_encoder):        
    # interpolate position embedding
    embedding_size = pos_embed_checkpoint.shape[-1]
    num_patches = visual_encoder.patch_embed.num_patches
    num_extra_tokens = visual_encoder.pos_embed.shape[-2] - num_patches
    # height (== width) for the checkpoint position embedding
    orig_size = int((pos_embed_checkpoint.shape[-2] - num_extra_tokens) ** 0.5)
    # height (== width) for the new position embedding
    new_size = int(num_patches ** 0.5)

    if orig_size!=new_size:
        # class_token and dist_token are kept unchanged
        extra_tokens = pos_embed_checkpoint[:, :num_extra_tokens]
        # only the position tokens are interpolated
        pos_tokens = pos_embed_checkpoint[:, num_extra_tokens:]
        pos_tokens = pos_tokens.reshape(-1, orig_size, orig_size, embedding_size).permute(0, 3, 1, 2)
        pos_tokens = torch.nn.functional.interpolate(
            pos_tokens, size=(new_size, new_size), mode='bicubic', align_corners=False)
        pos_tokens = pos_tokens.permute(0, 2, 3, 1).flatten(1, 2)
        new_pos_embed = torch.cat((extra_tokens, pos_tokens), dim=1)
        print('reshape position embedding from %d to %d'%(orig_size ** 2,new_size ** 2))
        
        return new_pos_embed    
    else:
        return pos_embed_checkpoint