File size: 29,686 Bytes
1ec58a5 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 |
# pytorch_diffusion + derived encoder decoder
import math
import numpy as np
import tqdm
import torch
import torch.nn as nn
from diffusers import DiffusionPipeline
from diffusers.configuration_utils import ConfigMixin
from diffusers.modeling_utils import ModelMixin
def get_timestep_embedding(timesteps, embedding_dim):
"""
This matches the implementation in Denoising Diffusion Probabilistic Models:
From Fairseq.
Build sinusoidal embeddings.
This matches the implementation in tensor2tensor, but differs slightly
from the description in Section 3.5 of "Attention Is All You Need".
"""
assert len(timesteps.shape) == 1
half_dim = embedding_dim // 2
emb = math.log(10000) / (half_dim - 1)
emb = torch.exp(torch.arange(half_dim, dtype=torch.float32) * -emb)
emb = emb.to(device=timesteps.device)
emb = timesteps.float()[:, None] * emb[None, :]
emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1)
if embedding_dim % 2 == 1: # zero pad
emb = torch.nn.functional.pad(emb, (0, 1, 0, 0))
return emb
def nonlinearity(x):
# swish
return x * torch.sigmoid(x)
def Normalize(in_channels):
return torch.nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True)
class Upsample(nn.Module):
def __init__(self, in_channels, with_conv):
super().__init__()
self.with_conv = with_conv
if self.with_conv:
self.conv = torch.nn.Conv2d(in_channels, in_channels, kernel_size=3, stride=1, padding=1)
def forward(self, x):
x = torch.nn.functional.interpolate(x, scale_factor=2.0, mode="nearest")
if self.with_conv:
x = self.conv(x)
return x
class Downsample(nn.Module):
def __init__(self, in_channels, with_conv):
super().__init__()
self.with_conv = with_conv
if self.with_conv:
# no asymmetric padding in torch conv, must do it ourselves
self.conv = torch.nn.Conv2d(in_channels, in_channels, kernel_size=3, stride=2, padding=0)
def forward(self, x):
if self.with_conv:
pad = (0, 1, 0, 1)
x = torch.nn.functional.pad(x, pad, mode="constant", value=0)
x = self.conv(x)
else:
x = torch.nn.functional.avg_pool2d(x, kernel_size=2, stride=2)
return x
class ResnetBlock(nn.Module):
def __init__(self, *, in_channels, out_channels=None, conv_shortcut=False, dropout, temb_channels=512):
super().__init__()
self.in_channels = in_channels
out_channels = in_channels if out_channels is None else out_channels
self.out_channels = out_channels
self.use_conv_shortcut = conv_shortcut
self.norm1 = Normalize(in_channels)
self.conv1 = torch.nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1)
if temb_channels > 0:
self.temb_proj = torch.nn.Linear(temb_channels, out_channels)
self.norm2 = Normalize(out_channels)
self.dropout = torch.nn.Dropout(dropout)
self.conv2 = torch.nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1)
if self.in_channels != self.out_channels:
if self.use_conv_shortcut:
self.conv_shortcut = torch.nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1)
else:
self.nin_shortcut = torch.nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=1, padding=0)
def forward(self, x, temb):
h = x
h = self.norm1(h)
h = nonlinearity(h)
h = self.conv1(h)
if temb is not None:
h = h + self.temb_proj(nonlinearity(temb))[:, :, None, None]
h = self.norm2(h)
h = nonlinearity(h)
h = self.dropout(h)
h = self.conv2(h)
if self.in_channels != self.out_channels:
if self.use_conv_shortcut:
x = self.conv_shortcut(x)
else:
x = self.nin_shortcut(x)
return x + h
class AttnBlock(nn.Module):
def __init__(self, in_channels):
super().__init__()
self.in_channels = in_channels
self.norm = Normalize(in_channels)
self.q = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
self.k = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
self.v = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
self.proj_out = torch.nn.Conv2d(in_channels, in_channels, kernel_size=1, stride=1, padding=0)
def forward(self, x):
h_ = x
h_ = self.norm(h_)
q = self.q(h_)
k = self.k(h_)
v = self.v(h_)
# compute attention
b, c, h, w = q.shape
q = q.reshape(b, c, h * w)
q = q.permute(0, 2, 1) # b,hw,c
k = k.reshape(b, c, h * w) # b,c,hw
w_ = torch.bmm(q, k) # b,hw,hw w[b,i,j]=sum_c q[b,i,c]k[b,c,j]
w_ = w_ * (int(c) ** (-0.5))
w_ = torch.nn.functional.softmax(w_, dim=2)
# attend to values
v = v.reshape(b, c, h * w)
w_ = w_.permute(0, 2, 1) # b,hw,hw (first hw of k, second of q)
h_ = torch.bmm(v, w_) # b, c,hw (hw of q) h_[b,c,j] = sum_i v[b,c,i] w_[b,i,j]
h_ = h_.reshape(b, c, h, w)
h_ = self.proj_out(h_)
return x + h_
class Model(nn.Module):
def __init__(
self,
*,
ch,
out_ch,
ch_mult=(1, 2, 4, 8),
num_res_blocks,
attn_resolutions,
dropout=0.0,
resamp_with_conv=True,
in_channels,
resolution,
use_timestep=True,
):
super().__init__()
self.ch = ch
self.temb_ch = self.ch * 4
self.num_resolutions = len(ch_mult)
self.num_res_blocks = num_res_blocks
self.resolution = resolution
self.in_channels = in_channels
self.use_timestep = use_timestep
if self.use_timestep:
# timestep embedding
self.temb = nn.Module()
self.temb.dense = nn.ModuleList(
[
torch.nn.Linear(self.ch, self.temb_ch),
torch.nn.Linear(self.temb_ch, self.temb_ch),
]
)
# downsampling
self.conv_in = torch.nn.Conv2d(in_channels, self.ch, kernel_size=3, stride=1, padding=1)
curr_res = resolution
in_ch_mult = (1,) + tuple(ch_mult)
self.down = nn.ModuleList()
for i_level in range(self.num_resolutions):
block = nn.ModuleList()
attn = nn.ModuleList()
block_in = ch * in_ch_mult[i_level]
block_out = ch * ch_mult[i_level]
for i_block in range(self.num_res_blocks):
block.append(
ResnetBlock(
in_channels=block_in, out_channels=block_out, temb_channels=self.temb_ch, dropout=dropout
)
)
block_in = block_out
if curr_res in attn_resolutions:
attn.append(AttnBlock(block_in))
down = nn.Module()
down.block = block
down.attn = attn
if i_level != self.num_resolutions - 1:
down.downsample = Downsample(block_in, resamp_with_conv)
curr_res = curr_res // 2
self.down.append(down)
# middle
self.mid = nn.Module()
self.mid.block_1 = ResnetBlock(
in_channels=block_in, out_channels=block_in, temb_channels=self.temb_ch, dropout=dropout
)
self.mid.attn_1 = AttnBlock(block_in)
self.mid.block_2 = ResnetBlock(
in_channels=block_in, out_channels=block_in, temb_channels=self.temb_ch, dropout=dropout
)
# upsampling
self.up = nn.ModuleList()
for i_level in reversed(range(self.num_resolutions)):
block = nn.ModuleList()
attn = nn.ModuleList()
block_out = ch * ch_mult[i_level]
skip_in = ch * ch_mult[i_level]
for i_block in range(self.num_res_blocks + 1):
if i_block == self.num_res_blocks:
skip_in = ch * in_ch_mult[i_level]
block.append(
ResnetBlock(
in_channels=block_in + skip_in,
out_channels=block_out,
temb_channels=self.temb_ch,
dropout=dropout,
)
)
block_in = block_out
if curr_res in attn_resolutions:
attn.append(AttnBlock(block_in))
up = nn.Module()
up.block = block
up.attn = attn
if i_level != 0:
up.upsample = Upsample(block_in, resamp_with_conv)
curr_res = curr_res * 2
self.up.insert(0, up) # prepend to get consistent order
# end
self.norm_out = Normalize(block_in)
self.conv_out = torch.nn.Conv2d(block_in, out_ch, kernel_size=3, stride=1, padding=1)
def forward(self, x, t=None):
# assert x.shape[2] == x.shape[3] == self.resolution
if self.use_timestep:
# timestep embedding
assert t is not None
temb = get_timestep_embedding(t, self.ch)
temb = self.temb.dense[0](temb)
temb = nonlinearity(temb)
temb = self.temb.dense[1](temb)
else:
temb = None
# downsampling
hs = [self.conv_in(x)]
for i_level in range(self.num_resolutions):
for i_block in range(self.num_res_blocks):
h = self.down[i_level].block[i_block](hs[-1], temb)
if len(self.down[i_level].attn) > 0:
h = self.down[i_level].attn[i_block](h)
hs.append(h)
if i_level != self.num_resolutions - 1:
hs.append(self.down[i_level].downsample(hs[-1]))
# middle
h = hs[-1]
h = self.mid.block_1(h, temb)
h = self.mid.attn_1(h)
h = self.mid.block_2(h, temb)
# upsampling
for i_level in reversed(range(self.num_resolutions)):
for i_block in range(self.num_res_blocks + 1):
h = self.up[i_level].block[i_block](torch.cat([h, hs.pop()], dim=1), temb)
if len(self.up[i_level].attn) > 0:
h = self.up[i_level].attn[i_block](h)
if i_level != 0:
h = self.up[i_level].upsample(h)
# end
h = self.norm_out(h)
h = nonlinearity(h)
h = self.conv_out(h)
return h
class Encoder(nn.Module):
def __init__(
self,
*,
ch,
out_ch,
ch_mult=(1, 2, 4, 8),
num_res_blocks,
attn_resolutions,
dropout=0.0,
resamp_with_conv=True,
in_channels,
resolution,
z_channels,
double_z=True,
**ignore_kwargs,
):
super().__init__()
self.ch = ch
self.temb_ch = 0
self.num_resolutions = len(ch_mult)
self.num_res_blocks = num_res_blocks
self.resolution = resolution
self.in_channels = in_channels
# downsampling
self.conv_in = torch.nn.Conv2d(in_channels, self.ch, kernel_size=3, stride=1, padding=1)
curr_res = resolution
in_ch_mult = (1,) + tuple(ch_mult)
self.down = nn.ModuleList()
for i_level in range(self.num_resolutions):
block = nn.ModuleList()
attn = nn.ModuleList()
block_in = ch * in_ch_mult[i_level]
block_out = ch * ch_mult[i_level]
for i_block in range(self.num_res_blocks):
block.append(
ResnetBlock(
in_channels=block_in, out_channels=block_out, temb_channels=self.temb_ch, dropout=dropout
)
)
block_in = block_out
if curr_res in attn_resolutions:
attn.append(AttnBlock(block_in))
down = nn.Module()
down.block = block
down.attn = attn
if i_level != self.num_resolutions - 1:
down.downsample = Downsample(block_in, resamp_with_conv)
curr_res = curr_res // 2
self.down.append(down)
# middle
self.mid = nn.Module()
self.mid.block_1 = ResnetBlock(
in_channels=block_in, out_channels=block_in, temb_channels=self.temb_ch, dropout=dropout
)
self.mid.attn_1 = AttnBlock(block_in)
self.mid.block_2 = ResnetBlock(
in_channels=block_in, out_channels=block_in, temb_channels=self.temb_ch, dropout=dropout
)
# end
self.norm_out = Normalize(block_in)
self.conv_out = torch.nn.Conv2d(
block_in, 2 * z_channels if double_z else z_channels, kernel_size=3, stride=1, padding=1
)
def forward(self, x):
# assert x.shape[2] == x.shape[3] == self.resolution, "{}, {}, {}".format(x.shape[2], x.shape[3], self.resolution)
# timestep embedding
temb = None
# downsampling
hs = [self.conv_in(x)]
for i_level in range(self.num_resolutions):
for i_block in range(self.num_res_blocks):
h = self.down[i_level].block[i_block](hs[-1], temb)
if len(self.down[i_level].attn) > 0:
h = self.down[i_level].attn[i_block](h)
hs.append(h)
if i_level != self.num_resolutions - 1:
hs.append(self.down[i_level].downsample(hs[-1]))
# middle
h = hs[-1]
h = self.mid.block_1(h, temb)
h = self.mid.attn_1(h)
h = self.mid.block_2(h, temb)
# end
h = self.norm_out(h)
h = nonlinearity(h)
h = self.conv_out(h)
return h
class Decoder(nn.Module):
def __init__(
self,
*,
ch,
out_ch,
ch_mult=(1, 2, 4, 8),
num_res_blocks,
attn_resolutions,
dropout=0.0,
resamp_with_conv=True,
in_channels,
resolution,
z_channels,
give_pre_end=False,
**ignorekwargs,
):
super().__init__()
self.ch = ch
self.temb_ch = 0
self.num_resolutions = len(ch_mult)
self.num_res_blocks = num_res_blocks
self.resolution = resolution
self.in_channels = in_channels
self.give_pre_end = give_pre_end
# compute in_ch_mult, block_in and curr_res at lowest res
in_ch_mult = (1,) + tuple(ch_mult)
block_in = ch * ch_mult[self.num_resolutions - 1]
curr_res = resolution // 2 ** (self.num_resolutions - 1)
self.z_shape = (1, z_channels, curr_res, curr_res)
print("Working with z of shape {} = {} dimensions.".format(self.z_shape, np.prod(self.z_shape)))
# z to block_in
self.conv_in = torch.nn.Conv2d(z_channels, block_in, kernel_size=3, stride=1, padding=1)
# middle
self.mid = nn.Module()
self.mid.block_1 = ResnetBlock(
in_channels=block_in, out_channels=block_in, temb_channels=self.temb_ch, dropout=dropout
)
self.mid.attn_1 = AttnBlock(block_in)
self.mid.block_2 = ResnetBlock(
in_channels=block_in, out_channels=block_in, temb_channels=self.temb_ch, dropout=dropout
)
# upsampling
self.up = nn.ModuleList()
for i_level in reversed(range(self.num_resolutions)):
block = nn.ModuleList()
attn = nn.ModuleList()
block_out = ch * ch_mult[i_level]
for i_block in range(self.num_res_blocks + 1):
block.append(
ResnetBlock(
in_channels=block_in, out_channels=block_out, temb_channels=self.temb_ch, dropout=dropout
)
)
block_in = block_out
if curr_res in attn_resolutions:
attn.append(AttnBlock(block_in))
up = nn.Module()
up.block = block
up.attn = attn
if i_level != 0:
up.upsample = Upsample(block_in, resamp_with_conv)
curr_res = curr_res * 2
self.up.insert(0, up) # prepend to get consistent order
# end
self.norm_out = Normalize(block_in)
self.conv_out = torch.nn.Conv2d(block_in, out_ch, kernel_size=3, stride=1, padding=1)
def forward(self, z):
# assert z.shape[1:] == self.z_shape[1:]
self.last_z_shape = z.shape
# timestep embedding
temb = None
# z to block_in
h = self.conv_in(z)
# middle
h = self.mid.block_1(h, temb)
h = self.mid.attn_1(h)
h = self.mid.block_2(h, temb)
# upsampling
for i_level in reversed(range(self.num_resolutions)):
for i_block in range(self.num_res_blocks + 1):
h = self.up[i_level].block[i_block](h, temb)
if len(self.up[i_level].attn) > 0:
h = self.up[i_level].attn[i_block](h)
if i_level != 0:
h = self.up[i_level].upsample(h)
# end
if self.give_pre_end:
return h
h = self.norm_out(h)
h = nonlinearity(h)
h = self.conv_out(h)
return h
class VectorQuantizer(nn.Module):
"""
Improved version over VectorQuantizer, can be used as a drop-in replacement. Mostly
avoids costly matrix multiplications and allows for post-hoc remapping of indices.
"""
# NOTE: due to a bug the beta term was applied to the wrong term. for
# backwards compatibility we use the buggy version by default, but you can
# specify legacy=False to fix it.
def __init__(self, n_e, e_dim, beta, remap=None, unknown_index="random", sane_index_shape=False, legacy=True):
super().__init__()
self.n_e = n_e
self.e_dim = e_dim
self.beta = beta
self.legacy = legacy
self.embedding = nn.Embedding(self.n_e, self.e_dim)
self.embedding.weight.data.uniform_(-1.0 / self.n_e, 1.0 / self.n_e)
self.remap = remap
if self.remap is not None:
self.register_buffer("used", torch.tensor(np.load(self.remap)))
self.re_embed = self.used.shape[0]
self.unknown_index = unknown_index # "random" or "extra" or integer
if self.unknown_index == "extra":
self.unknown_index = self.re_embed
self.re_embed = self.re_embed + 1
print(
f"Remapping {self.n_e} indices to {self.re_embed} indices. "
f"Using {self.unknown_index} for unknown indices."
)
else:
self.re_embed = n_e
self.sane_index_shape = sane_index_shape
def remap_to_used(self, inds):
ishape = inds.shape
assert len(ishape) > 1
inds = inds.reshape(ishape[0], -1)
used = self.used.to(inds)
match = (inds[:, :, None] == used[None, None, ...]).long()
new = match.argmax(-1)
unknown = match.sum(2) < 1
if self.unknown_index == "random":
new[unknown] = torch.randint(0, self.re_embed, size=new[unknown].shape).to(device=new.device)
else:
new[unknown] = self.unknown_index
return new.reshape(ishape)
def unmap_to_all(self, inds):
ishape = inds.shape
assert len(ishape) > 1
inds = inds.reshape(ishape[0], -1)
used = self.used.to(inds)
if self.re_embed > self.used.shape[0]: # extra token
inds[inds >= self.used.shape[0]] = 0 # simply set to zero
back = torch.gather(used[None, :][inds.shape[0] * [0], :], 1, inds)
return back.reshape(ishape)
def forward(self, z, temp=None, rescale_logits=False, return_logits=False):
assert temp is None or temp == 1.0, "Only for interface compatible with Gumbel"
assert rescale_logits == False, "Only for interface compatible with Gumbel"
assert return_logits == False, "Only for interface compatible with Gumbel"
# reshape z -> (batch, height, width, channel) and flatten
z = rearrange(z, "b c h w -> b h w c").contiguous()
z_flattened = z.view(-1, self.e_dim)
# distances from z to embeddings e_j (z - e)^2 = z^2 + e^2 - 2 e * z
d = (
torch.sum(z_flattened**2, dim=1, keepdim=True)
+ torch.sum(self.embedding.weight**2, dim=1)
- 2 * torch.einsum("bd,dn->bn", z_flattened, rearrange(self.embedding.weight, "n d -> d n"))
)
min_encoding_indices = torch.argmin(d, dim=1)
z_q = self.embedding(min_encoding_indices).view(z.shape)
perplexity = None
min_encodings = None
# compute loss for embedding
if not self.legacy:
loss = self.beta * torch.mean((z_q.detach() - z) ** 2) + torch.mean((z_q - z.detach()) ** 2)
else:
loss = torch.mean((z_q.detach() - z) ** 2) + self.beta * torch.mean((z_q - z.detach()) ** 2)
# preserve gradients
z_q = z + (z_q - z).detach()
# reshape back to match original input shape
z_q = rearrange(z_q, "b h w c -> b c h w").contiguous()
if self.remap is not None:
min_encoding_indices = min_encoding_indices.reshape(z.shape[0], -1) # add batch axis
min_encoding_indices = self.remap_to_used(min_encoding_indices)
min_encoding_indices = min_encoding_indices.reshape(-1, 1) # flatten
if self.sane_index_shape:
min_encoding_indices = min_encoding_indices.reshape(z_q.shape[0], z_q.shape[2], z_q.shape[3])
return z_q, loss, (perplexity, min_encodings, min_encoding_indices)
def get_codebook_entry(self, indices, shape):
# shape specifying (batch, height, width, channel)
if self.remap is not None:
indices = indices.reshape(shape[0], -1) # add batch axis
indices = self.unmap_to_all(indices)
indices = indices.reshape(-1) # flatten again
# get quantized latent vectors
z_q = self.embedding(indices)
if shape is not None:
z_q = z_q.view(shape)
# reshape back to match original input shape
z_q = z_q.permute(0, 3, 1, 2).contiguous()
return z_q
class VQModel(ModelMixin, ConfigMixin):
def __init__(
self,
ch,
out_ch,
num_res_blocks,
attn_resolutions,
in_channels,
resolution,
z_channels,
n_embed,
embed_dim,
remap=None,
sane_index_shape=False, # tell vector quantizer to return indices as bhw
ch_mult=(1, 2, 4, 8),
dropout=0.0,
double_z=True,
resamp_with_conv=True,
give_pre_end=False,
):
super().__init__()
# register all __init__ params with self.register
self.register(
ch=ch,
out_ch=out_ch,
num_res_blocks=num_res_blocks,
attn_resolutions=attn_resolutions,
in_channels=in_channels,
resolution=resolution,
z_channels=z_channels,
n_embed=n_embed,
embed_dim=embed_dim,
remap=remap,
sane_index_shape=sane_index_shape,
ch_mult=ch_mult,
dropout=dropout,
double_z=double_z,
resamp_with_conv=resamp_with_conv,
give_pre_end=give_pre_end,
)
# pass init params to Encoder
self.encoder = Encoder(
ch=ch,
out_ch=out_ch,
num_res_blocks=num_res_blocks,
attn_resolutions=attn_resolutions,
in_channels=in_channels,
resolution=resolution,
z_channels=z_channels,
ch_mult=ch_mult,
dropout=dropout,
resamp_with_conv=resamp_with_conv,
double_z=double_z,
give_pre_end=give_pre_end,
)
self.quantize = VectorQuantizer(n_embed, embed_dim, beta=0.25, remap=remap, sane_index_shape=sane_index_shape)
# pass init params to Decoder
self.decoder = Decoder(
ch=ch,
out_ch=out_ch,
num_res_blocks=num_res_blocks,
attn_resolutions=attn_resolutions,
in_channels=in_channels,
resolution=resolution,
z_channels=z_channels,
ch_mult=ch_mult,
dropout=dropout,
resamp_with_conv=resamp_with_conv,
give_pre_end=give_pre_end,
)
def encode(self, x):
h = self.encoder(x)
h = self.quant_conv(h)
return h
def decode(self, h, force_not_quantize=False):
# also go through quantization layer
if not force_not_quantize:
quant, emb_loss, info = self.quantize(h)
else:
quant = h
quant = self.post_quant_conv(quant)
dec = self.decoder(quant)
return dec
class DiagonalGaussianDistribution(object):
def __init__(self, parameters, deterministic=False):
self.parameters = parameters
self.mean, self.logvar = torch.chunk(parameters, 2, dim=1)
self.logvar = torch.clamp(self.logvar, -30.0, 20.0)
self.deterministic = deterministic
self.std = torch.exp(0.5 * self.logvar)
self.var = torch.exp(self.logvar)
if self.deterministic:
self.var = self.std = torch.zeros_like(self.mean).to(device=self.parameters.device)
def sample(self):
x = self.mean + self.std * torch.randn(self.mean.shape).to(device=self.parameters.device)
return x
def kl(self, other=None):
if self.deterministic:
return torch.Tensor([0.])
else:
if other is None:
return 0.5 * torch.sum(torch.pow(self.mean, 2)
+ self.var - 1.0 - self.logvar,
dim=[1, 2, 3])
else:
return 0.5 * torch.sum(
torch.pow(self.mean - other.mean, 2) / other.var
+ self.var / other.var - 1.0 - self.logvar + other.logvar,
dim=[1, 2, 3])
def nll(self, sample, dims=[1,2,3]):
if self.deterministic:
return torch.Tensor([0.])
logtwopi = np.log(2.0 * np.pi)
return 0.5 * torch.sum(
logtwopi + self.logvar + torch.pow(sample - self.mean, 2) / self.var,
dim=dims)
def mode(self):
return self.mean
class AutoencoderKL(ModelMixin, ConfigMixin):
def __init__(
self,
ch,
out_ch,
num_res_blocks,
attn_resolutions,
in_channels,
resolution,
z_channels,
embed_dim,
remap=None,
sane_index_shape=False, # tell vector quantizer to return indices as bhw
ch_mult=(1, 2, 4, 8),
dropout=0.0,
double_z=True,
resamp_with_conv=True,
give_pre_end=False,
):
super().__init__()
# register all __init__ params with self.register
self.register(
ch=ch,
out_ch=out_ch,
num_res_blocks=num_res_blocks,
attn_resolutions=attn_resolutions,
in_channels=in_channels,
resolution=resolution,
z_channels=z_channels,
embed_dim=embed_dim,
remap=remap,
sane_index_shape=sane_index_shape,
ch_mult=ch_mult,
dropout=dropout,
double_z=double_z,
resamp_with_conv=resamp_with_conv,
give_pre_end=give_pre_end,
)
# pass init params to Encoder
self.encoder = Encoder(
ch=ch,
out_ch=out_ch,
num_res_blocks=num_res_blocks,
attn_resolutions=attn_resolutions,
in_channels=in_channels,
resolution=resolution,
z_channels=z_channels,
ch_mult=ch_mult,
dropout=dropout,
resamp_with_conv=resamp_with_conv,
double_z=double_z,
give_pre_end=give_pre_end,
)
# pass init params to Decoder
self.decoder = Decoder(
ch=ch,
out_ch=out_ch,
num_res_blocks=num_res_blocks,
attn_resolutions=attn_resolutions,
in_channels=in_channels,
resolution=resolution,
z_channels=z_channels,
ch_mult=ch_mult,
dropout=dropout,
resamp_with_conv=resamp_with_conv,
give_pre_end=give_pre_end,
)
self.quant_conv = torch.nn.Conv2d(2*z_channels, 2*embed_dim, 1)
self.post_quant_conv = torch.nn.Conv2d(embed_dim, z_channels, 1)
def encode(self, x):
h = self.encoder(x)
moments = self.quant_conv(h)
posterior = DiagonalGaussianDistribution(moments)
return posterior
def decode(self, z):
z = self.post_quant_conv(z)
dec = self.decoder(z)
return dec
def forward(self, input, sample_posterior=True):
posterior = self.encode(input)
if sample_posterior:
z = posterior.sample()
else:
z = posterior.mode()
dec = self.decode(z)
return dec, posterior |