-
Notifications
You must be signed in to change notification settings - Fork 1
/
models.py
352 lines (272 loc) · 13.4 KB
/
models.py
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
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
import torch
import torch.nn as nn
import numpy as np
import math
from timm.models.vision_transformer import PatchEmbed, Attention, Mlp
# def modulate(x, shift, scale):
# return x * (1 + scale.unsqueeze(1)) + shift.unsqueeze(1)
def modulate(x, shift, scale):
return x * (1 + scale) + shift
class TimestepEmbedder(nn.Module):
"""
Embeds scalar timesteps into vector representations.
"""
def __init__(self, input_size, patch_size, in_channels, hidden_size, frequency_embedding_size=256):
# (self, hidden_size, frequency_embedding_size=256):
compress_size = 4
super().__init__()
self.mlp = nn.Sequential(
nn.Linear(frequency_embedding_size, compress_size, bias=True),
nn.SiLU(),
nn.Linear(compress_size, compress_size, bias=True),
)
self.frequency_embedding_size = frequency_embedding_size
self.patchEmb = PatchEmbed(input_size, patch_size, in_channels*compress_size, hidden_size, norm_layer=None, bias=True)
@staticmethod
def timestep_embedding(t, dim, max_period=10000):
"""
Create sinusoidal timestep embeddings.
:param t: 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, D) Tensor of positional embeddings.
"""
half = dim // 2
freqs = torch.exp(
-math.log(max_period) * torch.arange(start=0, end=half, dtype=torch.float32) / half
).to(device=t.device)
args = t[..., None].float() * freqs[None]
embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
if dim % 2:
embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1)
return embedding
def forward(self, t):
t_freq = self.timestep_embedding(t, self.frequency_embedding_size)
t_freq = self.mlp(t_freq)
t_freq = torch.einsum('nchwf->ncfhw',t_freq)
t_freq = t_freq.reshape(t_freq.shape[0], -1, t_freq.shape[3], t_freq.shape[4])
t_emb = self.patchEmb(t_freq)
return t_emb
class LabelEmbedder(nn.Module):
"""
Embeds class labels into vector representations. Also handles label dropout for classifier-free guidance.
"""
def __init__(self, num_classes, hidden_size, dropout_prob):
super().__init__()
use_cfg_embedding = dropout_prob > 0
self.embedding_table = nn.Embedding(num_classes + use_cfg_embedding, hidden_size)
self.num_classes = num_classes
self.dropout_prob = dropout_prob
def token_drop(self, labels, force_drop_ids=None):
"""
Drops labels to enable classifier-free guidance.
"""
if force_drop_ids is None:
drop_ids = torch.rand(labels.shape[0], device=labels.device) < self.dropout_prob
else:
drop_ids = force_drop_ids == 1
labels = torch.where(drop_ids, self.num_classes, labels)
return labels
def forward(self, labels, train, force_drop_ids=None):
use_dropout = self.dropout_prob > 0
if (train and use_dropout) or (force_drop_ids is not None):
labels = self.token_drop(labels, force_drop_ids)
embeddings = self.embedding_table(labels)
return embeddings
class ComboStocBlock(nn.Module):
def __init__(self, hidden_size, num_heads, mlp_ratio=4.0, **block_kwargs):
super().__init__()
self.norm1 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
self.attn = Attention(hidden_size, num_heads=num_heads, qkv_bias=True, **block_kwargs)
self.norm2 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
mlp_hidden_dim = int(hidden_size * mlp_ratio)
approx_gelu = lambda: nn.GELU(approximate="tanh")
self.mlp = Mlp(in_features=hidden_size, hidden_features=mlp_hidden_dim, act_layer=approx_gelu, drop=0)
self.adaLN_modulation = nn.Sequential(
nn.SiLU(),
nn.Linear(hidden_size, 6 * hidden_size, bias=True)
)
def forward(self, x, c):
shift_msa, scale_msa, gate_msa, shift_mlp, scale_mlp, gate_mlp = self.adaLN_modulation(c).chunk(6, dim=-1)
x = x + gate_msa * self.attn(modulate(self.norm1(x), shift_msa, scale_msa))
x = x + gate_mlp * self.mlp(modulate(self.norm2(x), shift_mlp, scale_mlp))
return x
class FinalLayer(nn.Module):
def __init__(self, hidden_size, patch_size, out_channels):
super().__init__()
self.norm_final = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
self.linear = nn.Linear(hidden_size, patch_size * patch_size * out_channels, bias=True)
self.adaLN_modulation = nn.Sequential(
nn.SiLU(),
nn.Linear(hidden_size, 2 * hidden_size, bias=True)
)
def forward(self, x, c):
shift, scale = self.adaLN_modulation(c).chunk(2, dim=-1)
x = modulate(self.norm_final(x), shift, scale)
x = self.linear(x)
return x
class ComboStoc(nn.Module):
def __init__(
self,
input_size=32,
patch_size=2,
in_channels=4,
hidden_size=1152,
depth=28,
num_heads=16,
mlp_ratio=4.0,
class_dropout_prob=0.1,
num_classes=1000,
learn_sigma=True,
):
super().__init__()
self.learn_sigma = learn_sigma
self.in_channels = in_channels
self.out_channels = in_channels * 2 if learn_sigma else in_channels
self.patch_size = patch_size
self.num_heads = num_heads
self.x_embedder = PatchEmbed(input_size, patch_size, in_channels, hidden_size, bias=True)
# self.t_embedder = TimestepEmbedder(hidden_size)
self.t_embedder = TimestepEmbedder(input_size, patch_size, in_channels, hidden_size, frequency_embedding_size = 256)
self.y_embedder = LabelEmbedder(num_classes, hidden_size, class_dropout_prob)
num_patches = self.x_embedder.num_patches
# Will use fixed sin-cos embedding:
self.pos_embed = nn.Parameter(torch.zeros(1, num_patches, hidden_size), requires_grad=False)
self.blocks = nn.ModuleList([
ComboStocBlock(hidden_size, num_heads, mlp_ratio=mlp_ratio) for _ in range(depth)
])
self.final_layer = FinalLayer(hidden_size, patch_size, self.out_channels)
self.initialize_weights()
def initialize_weights(self):
# Initialize transformer layers:
def _basic_init(module):
if isinstance(module, nn.Linear):
torch.nn.init.xavier_uniform_(module.weight)
if module.bias is not None:
nn.init.constant_(module.bias, 0)
self.apply(_basic_init)
# Initialize (and freeze) pos_embed by sin-cos embedding:
pos_embed = get_2d_sincos_pos_embed(self.pos_embed.shape[-1], int(self.x_embedder.num_patches ** 0.5))
self.pos_embed.data.copy_(torch.from_numpy(pos_embed).float().unsqueeze(0))
# Initialize patch_embed like nn.Linear (instead of nn.Conv2d):
w = self.x_embedder.proj.weight.data
nn.init.xavier_uniform_(w.view([w.shape[0], -1]))
nn.init.constant_(self.x_embedder.proj.bias, 0)
# Initialize label embedding table:
nn.init.normal_(self.y_embedder.embedding_table.weight, std=0.02)
# Initialize timestep embedding MLP:
nn.init.normal_(self.t_embedder.mlp[0].weight, std=0.02)
nn.init.normal_(self.t_embedder.mlp[2].weight, std=0.02)
w = self.t_embedder.patchEmb.proj.weight.data
nn.init.xavier_uniform_(w.view([w.shape[0], -1]))
nn.init.constant_(self.t_embedder.patchEmb.proj.bias, 0)
# Zero-out adaLN modulation layers in ComboStoc blocks:
for block in self.blocks:
nn.init.constant_(block.adaLN_modulation[-1].weight, 0)
nn.init.constant_(block.adaLN_modulation[-1].bias, 0)
# Zero-out output layers:
nn.init.constant_(self.final_layer.adaLN_modulation[-1].weight, 0)
nn.init.constant_(self.final_layer.adaLN_modulation[-1].bias, 0)
nn.init.constant_(self.final_layer.linear.weight, 0)
nn.init.constant_(self.final_layer.linear.bias, 0)
def unpatchify(self, x):
c = self.out_channels
p = self.x_embedder.patch_size[0]
h = w = int(x.shape[1] ** 0.5)
assert h * w == x.shape[1]
x = x.reshape(shape=(x.shape[0], h, w, p, p, c))
x = torch.einsum('nhwpqc->nchpwq', x)
imgs = x.reshape(shape=(x.shape[0], c, h * p, h * p))
return imgs
def ckpt_wrapper(self, module):
def ckpt_forward(*inputs):
outputs = module(*inputs)
return outputs
return ckpt_forward
def forward(self, x, t, y):
x_size = x.size()
x = self.x_embedder(x) + self.pos_embed # (N, T, D), where T = H * W / patch_size ** 2
if t.shape != x_size: # if t is not of shape x, expand it. haopan
dims = [1] * (len(x_size) - 1)
t = t.view(t.size(0), *dims)
t = t.repeat(1, x_size[1], x_size[2], x_size[3])
t = self.t_embedder(t) # (N, T, D)
y = self.y_embedder(y, self.training) # (N, D)
y = y.unsqueeze(1) # (N, 1, D)
c = t + y # (N, T, D)
for block in self.blocks:
# x = block(x, c) # (N, T, D)
x = torch.utils.checkpoint.checkpoint(self.ckpt_wrapper(block), x, c, use_reentrant=False) # (N, T, D)
x = self.final_layer(x, c) # (N, T, patch_size ** 2 * out_channels)
x = self.unpatchify(x) # (N, out_channels, H, W)
if self.learn_sigma:
x, _ = x.chunk(2, dim=1)
return x
def forward_with_cfg(self, x, t, y, cfg_scale):
half = x[: len(x) // 2]
combined = torch.cat([half, half], dim=0)
model_out = self.forward(combined, t, y)
eps, rest = model_out[:, :3], model_out[:, 3:]
cond_eps, uncond_eps = torch.split(eps, len(eps) // 2, dim=0)
half_eps = uncond_eps + cfg_scale * (cond_eps - uncond_eps)
eps = torch.cat([half_eps, half_eps], dim=0)
return torch.cat([eps, rest], dim=1)
def get_2d_sincos_pos_embed(embed_dim, grid_size, cls_token=False, extra_tokens=0):
grid_h = np.arange(grid_size, dtype=np.float32)
grid_w = np.arange(grid_size, dtype=np.float32)
grid = np.meshgrid(grid_w, grid_h) # here w goes first
grid = np.stack(grid, axis=0)
grid = grid.reshape([2, 1, grid_size, grid_size])
pos_embed = get_2d_sincos_pos_embed_from_grid(embed_dim, grid)
if cls_token and extra_tokens > 0:
pos_embed = np.concatenate([np.zeros([extra_tokens, embed_dim]), pos_embed], axis=0)
return pos_embed
def get_2d_sincos_pos_embed_from_grid(embed_dim, grid):
assert embed_dim % 2 == 0
# use half of dimensions to encode grid_h
emb_h = get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[0]) # (H*W, D/2)
emb_w = get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[1]) # (H*W, D/2)
emb = np.concatenate([emb_h, emb_w], axis=1) # (H*W, D)
return emb
def get_1d_sincos_pos_embed_from_grid(embed_dim, pos):
assert embed_dim % 2 == 0
omega = np.arange(embed_dim // 2, dtype=np.float64)
omega /= embed_dim / 2.
omega = 1. / 10000**omega # (D/2,)
pos = pos.reshape(-1) # (M,)
out = np.einsum('m,d->md', pos, omega) # (M, D/2), outer product
emb_sin = np.sin(out) # (M, D/2)
emb_cos = np.cos(out) # (M, D/2)
emb = np.concatenate([emb_sin, emb_cos], axis=1) # (M, D)
return emb
def ComboStoc_XL_2(**kwargs):
return ComboStoc(depth=28, hidden_size=1152, patch_size=2, num_heads=16, **kwargs)
def ComboStoc_XL_4(**kwargs):
return ComboStoc(depth=28, hidden_size=1152, patch_size=4, num_heads=16, **kwargs)
def ComboStoc_XL_8(**kwargs):
return ComboStoc(depth=28, hidden_size=1152, patch_size=8, num_heads=16, **kwargs)
def ComboStoc_L_2(**kwargs):
return ComboStoc(depth=24, hidden_size=1024, patch_size=2, num_heads=16, **kwargs)
def ComboStoc_L_4(**kwargs):
return ComboStoc(depth=24, hidden_size=1024, patch_size=4, num_heads=16, **kwargs)
def ComboStoc_L_8(**kwargs):
return ComboStoc(depth=24, hidden_size=1024, patch_size=8, num_heads=16, **kwargs)
def ComboStoc_B_2(**kwargs):
return ComboStoc(depth=12, hidden_size=768, patch_size=2, num_heads=12, **kwargs)
def ComboStoc_B_4(**kwargs):
return ComboStoc(depth=12, hidden_size=768, patch_size=4, num_heads=12, **kwargs)
def ComboStoc_B_8(**kwargs):
return ComboStoc(depth=12, hidden_size=768, patch_size=8, num_heads=12, **kwargs)
def ComboStoc_S_2(**kwargs):
return ComboStoc(depth=12, hidden_size=384, patch_size=2, num_heads=6, **kwargs)
def ComboStoc_S_4(**kwargs):
return ComboStoc(depth=12, hidden_size=384, patch_size=4, num_heads=6, **kwargs)
def ComboStoc_S_8(**kwargs):
return ComboStoc(depth=12, hidden_size=384, patch_size=8, num_heads=6, **kwargs)
ComboStoc_models = {
'ComboStoc-XL/2': ComboStoc_XL_2, 'ComboStoc-XL/4': ComboStoc_XL_4, 'ComboStoc-XL/8': ComboStoc_XL_8,
'ComboStoc-L/2': ComboStoc_L_2, 'ComboStoc-L/4': ComboStoc_L_4, 'ComboStoc-L/8': ComboStoc_L_8,
'ComboStoc-B/2': ComboStoc_B_2, 'ComboStoc-B/4': ComboStoc_B_4, 'ComboStoc-B/8': ComboStoc_B_8,
'ComboStoc-S/2': ComboStoc_S_2, 'ComboStoc-S/4': ComboStoc_S_4, 'ComboStoc-S/8': ComboStoc_S_8,
}