InvokeAI/invokeai/backend/restoration/vqgan_arch.py

515 lines
16 KiB
Python
Raw Normal View History

2023-03-03 06:02:00 +00:00
"""
VQGAN code, adapted from the original created by the Unleashing Transformers authors:
https://github.com/samb-t/unleashing-transformers/blob/master/models/vqgan.py
2023-03-03 06:02:00 +00:00
"""
import copy
import numpy as np
import torch
import torch.nn as nn
import torch.nn.functional as F
from basicsr.utils import get_root_logger
from basicsr.utils.registry import ARCH_REGISTRY
2023-03-03 06:02:00 +00:00
def normalize(in_channels):
2023-03-03 06:02:00 +00:00
return torch.nn.GroupNorm(
num_groups=32, num_channels=in_channels, eps=1e-6, affine=True
)
@torch.jit.script
def swish(x):
2023-03-03 06:02:00 +00:00
return x * torch.sigmoid(x)
# Define VQVAE classes
class VectorQuantizer(nn.Module):
def __init__(self, codebook_size, emb_dim, beta):
super(VectorQuantizer, self).__init__()
self.codebook_size = codebook_size # number of embeddings
self.emb_dim = emb_dim # dimension of embedding
self.beta = beta # commitment cost used in loss term, beta * ||z_e(x)-sg[e]||^2
self.embedding = nn.Embedding(self.codebook_size, self.emb_dim)
2023-03-03 06:02:00 +00:00
self.embedding.weight.data.uniform_(
-1.0 / self.codebook_size, 1.0 / self.codebook_size
)
def forward(self, z):
# reshape z -> (batch, height, width, channel) and flatten
z = z.permute(0, 2, 3, 1).contiguous()
z_flattened = z.view(-1, self.emb_dim)
# distances from z to embeddings e_j (z - e)^2 = z^2 + e^2 - 2 e * z
2023-03-03 06:02:00 +00:00
d = (
(z_flattened**2).sum(dim=1, keepdim=True)
+ (self.embedding.weight**2).sum(1)
- 2 * torch.matmul(z_flattened, self.embedding.weight.t())
)
mean_distance = torch.mean(d)
# find closest encodings
# min_encoding_indices = torch.argmin(d, dim=1).unsqueeze(1)
2023-03-03 06:02:00 +00:00
min_encoding_scores, min_encoding_indices = torch.topk(
d, 1, dim=1, largest=False
)
# [0-1], higher score, higher confidence
2023-03-03 06:02:00 +00:00
min_encoding_scores = torch.exp(-min_encoding_scores / 10)
2023-03-03 06:02:00 +00:00
min_encodings = torch.zeros(
min_encoding_indices.shape[0], self.codebook_size
).to(z)
min_encodings.scatter_(1, min_encoding_indices, 1)
# get quantized latent vectors
z_q = torch.matmul(min_encodings, self.embedding.weight).view(z.shape)
# compute loss for embedding
2023-03-03 06:02:00 +00:00
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()
# perplexity
e_mean = torch.mean(min_encodings, dim=0)
perplexity = torch.exp(-torch.sum(e_mean * torch.log(e_mean + 1e-10)))
# reshape back to match original input shape
z_q = z_q.permute(0, 3, 1, 2).contiguous()
2023-03-03 06:02:00 +00:00
return (
z_q,
loss,
{
"perplexity": perplexity,
"min_encodings": min_encodings,
"min_encoding_indices": min_encoding_indices,
"min_encoding_scores": min_encoding_scores,
"mean_distance": mean_distance,
},
)
def get_codebook_feat(self, indices, shape):
# input indices: batch*token_num -> (batch*token_num)*1
# shape: batch, height, width, channel
2023-03-03 06:02:00 +00:00
indices = indices.view(-1, 1)
min_encodings = torch.zeros(indices.shape[0], self.codebook_size).to(indices)
min_encodings.scatter_(1, indices, 1)
# get quantized latent vectors
z_q = torch.matmul(min_encodings.float(), self.embedding.weight)
if shape is not None: # reshape back to match original input shape
z_q = z_q.view(shape).permute(0, 3, 1, 2).contiguous()
return z_q
class GumbelQuantizer(nn.Module):
2023-03-03 06:02:00 +00:00
def __init__(
self,
codebook_size,
emb_dim,
num_hiddens,
straight_through=False,
kl_weight=5e-4,
temp_init=1.0,
):
super().__init__()
self.codebook_size = codebook_size # number of embeddings
self.emb_dim = emb_dim # dimension of embedding
self.straight_through = straight_through
self.temperature = temp_init
self.kl_weight = kl_weight
2023-03-03 06:02:00 +00:00
self.proj = nn.Conv2d(
num_hiddens, codebook_size, 1
) # projects last encoder layer to quantized logits
self.embed = nn.Embedding(codebook_size, emb_dim)
def forward(self, z):
hard = self.straight_through if self.training else True
logits = self.proj(z)
soft_one_hot = F.gumbel_softmax(logits, tau=self.temperature, dim=1, hard=hard)
z_q = torch.einsum("b n h w, n d -> b d h w", soft_one_hot, self.embed.weight)
# + kl divergence to the prior loss
qy = F.softmax(logits, dim=1)
2023-03-03 06:02:00 +00:00
diff = (
self.kl_weight
* torch.sum(qy * torch.log(qy * self.codebook_size + 1e-10), dim=1).mean()
)
min_encoding_indices = soft_one_hot.argmax(dim=1)
2023-03-03 06:02:00 +00:00
return z_q, diff, {"min_encoding_indices": min_encoding_indices}
class Downsample(nn.Module):
def __init__(self, in_channels):
super().__init__()
2023-03-03 06:02:00 +00:00
self.conv = torch.nn.Conv2d(
in_channels, in_channels, kernel_size=3, stride=2, padding=0
)
def forward(self, x):
pad = (0, 1, 0, 1)
x = torch.nn.functional.pad(x, pad, mode="constant", value=0)
x = self.conv(x)
return x
class Upsample(nn.Module):
def __init__(self, in_channels):
super().__init__()
2023-03-03 06:02:00 +00:00
self.conv = nn.Conv2d(
in_channels, in_channels, kernel_size=3, stride=1, padding=1
)
def forward(self, x):
x = F.interpolate(x, scale_factor=2.0, mode="nearest")
x = self.conv(x)
return x
class ResBlock(nn.Module):
def __init__(self, in_channels, out_channels=None):
super(ResBlock, self).__init__()
self.in_channels = in_channels
self.out_channels = in_channels if out_channels is None else out_channels
self.norm1 = normalize(in_channels)
2023-03-03 06:02:00 +00:00
self.conv1 = nn.Conv2d(
in_channels, out_channels, kernel_size=3, stride=1, padding=1
)
self.norm2 = normalize(out_channels)
2023-03-03 06:02:00 +00:00
self.conv2 = nn.Conv2d(
out_channels, out_channels, kernel_size=3, stride=1, padding=1
)
if self.in_channels != self.out_channels:
2023-03-03 06:02:00 +00:00
self.conv_out = nn.Conv2d(
in_channels, out_channels, kernel_size=1, stride=1, padding=0
)
def forward(self, x_in):
x = x_in
x = self.norm1(x)
x = swish(x)
x = self.conv1(x)
x = self.norm2(x)
x = swish(x)
x = self.conv2(x)
if self.in_channels != self.out_channels:
x_in = self.conv_out(x_in)
return x + x_in
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(
2023-03-03 06:02:00 +00:00
in_channels, in_channels, kernel_size=1, stride=1, padding=0
)
self.k = torch.nn.Conv2d(
2023-03-03 06:02:00 +00:00
in_channels, in_channels, kernel_size=1, stride=1, padding=0
)
self.v = torch.nn.Conv2d(
2023-03-03 06:02:00 +00:00
in_channels, in_channels, kernel_size=1, stride=1, padding=0
)
self.proj_out = torch.nn.Conv2d(
2023-03-03 06:02:00 +00:00
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
2023-03-03 06:02:00 +00:00
q = q.reshape(b, c, h * w)
q = q.permute(0, 2, 1)
2023-03-03 06:02:00 +00:00
k = k.reshape(b, c, h * w)
w_ = torch.bmm(q, k)
2023-03-03 06:02:00 +00:00
w_ = w_ * (int(c) ** (-0.5))
w_ = F.softmax(w_, dim=2)
# attend to values
2023-03-03 06:02:00 +00:00
v = v.reshape(b, c, h * w)
w_ = w_.permute(0, 2, 1)
h_ = torch.bmm(v, w_)
h_ = h_.reshape(b, c, h, w)
h_ = self.proj_out(h_)
2023-03-03 06:02:00 +00:00
return x + h_
class Encoder(nn.Module):
2023-03-03 06:02:00 +00:00
def __init__(
self,
in_channels,
nf,
emb_dim,
ch_mult,
num_res_blocks,
resolution,
attn_resolutions,
):
super().__init__()
self.nf = nf
self.num_resolutions = len(ch_mult)
self.num_res_blocks = num_res_blocks
self.resolution = resolution
self.attn_resolutions = attn_resolutions
curr_res = self.resolution
2023-03-03 06:02:00 +00:00
in_ch_mult = (1,) + tuple(ch_mult)
blocks = []
# initial convultion
blocks.append(nn.Conv2d(in_channels, nf, kernel_size=3, stride=1, padding=1))
# residual and downsampling blocks, with attention on smaller res (16x16)
for i in range(self.num_resolutions):
block_in_ch = nf * in_ch_mult[i]
block_out_ch = nf * ch_mult[i]
for _ in range(self.num_res_blocks):
blocks.append(ResBlock(block_in_ch, block_out_ch))
block_in_ch = block_out_ch
if curr_res in attn_resolutions:
blocks.append(AttnBlock(block_in_ch))
if i != self.num_resolutions - 1:
blocks.append(Downsample(block_in_ch))
curr_res = curr_res // 2
# non-local attention block
blocks.append(ResBlock(block_in_ch, block_in_ch))
blocks.append(AttnBlock(block_in_ch))
blocks.append(ResBlock(block_in_ch, block_in_ch))
# normalise and convert to latent size
blocks.append(normalize(block_in_ch))
2023-03-03 06:02:00 +00:00
blocks.append(
nn.Conv2d(block_in_ch, emb_dim, kernel_size=3, stride=1, padding=1)
)
self.blocks = nn.ModuleList(blocks)
def forward(self, x):
for block in self.blocks:
x = block(x)
return x
class Generator(nn.Module):
def __init__(self, nf, emb_dim, ch_mult, res_blocks, img_size, attn_resolutions):
super().__init__()
self.nf = nf
self.ch_mult = ch_mult
self.num_resolutions = len(self.ch_mult)
self.num_res_blocks = res_blocks
self.resolution = img_size
self.attn_resolutions = attn_resolutions
self.in_channels = emb_dim
self.out_channels = 3
block_in_ch = self.nf * self.ch_mult[-1]
2023-03-03 06:02:00 +00:00
curr_res = self.resolution // 2 ** (self.num_resolutions - 1)
blocks = []
# initial conv
2023-03-03 06:02:00 +00:00
blocks.append(
nn.Conv2d(self.in_channels, block_in_ch, kernel_size=3, stride=1, padding=1)
)
# non-local attention block
blocks.append(ResBlock(block_in_ch, block_in_ch))
blocks.append(AttnBlock(block_in_ch))
blocks.append(ResBlock(block_in_ch, block_in_ch))
for i in reversed(range(self.num_resolutions)):
block_out_ch = self.nf * self.ch_mult[i]
for _ in range(self.num_res_blocks):
blocks.append(ResBlock(block_in_ch, block_out_ch))
block_in_ch = block_out_ch
if curr_res in self.attn_resolutions:
blocks.append(AttnBlock(block_in_ch))
if i != 0:
blocks.append(Upsample(block_in_ch))
curr_res = curr_res * 2
blocks.append(normalize(block_in_ch))
2023-03-03 06:02:00 +00:00
blocks.append(
nn.Conv2d(
block_in_ch, self.out_channels, kernel_size=3, stride=1, padding=1
)
)
self.blocks = nn.ModuleList(blocks)
def forward(self, x):
for block in self.blocks:
x = block(x)
return x
@ARCH_REGISTRY.register()
class VQAutoEncoder(nn.Module):
2023-03-03 06:02:00 +00:00
def __init__(
self,
img_size,
nf,
ch_mult,
quantizer="nearest",
res_blocks=2,
attn_resolutions=[16],
codebook_size=1024,
emb_dim=256,
beta=0.25,
gumbel_straight_through=False,
gumbel_kl_weight=1e-8,
model_path=None,
):
super().__init__()
logger = get_root_logger()
self.in_channels = 3
self.nf = nf
self.n_blocks = res_blocks
self.codebook_size = codebook_size
self.embed_dim = emb_dim
self.ch_mult = ch_mult
self.resolution = img_size
self.attn_resolutions = attn_resolutions
self.quantizer_type = quantizer
self.encoder = Encoder(
self.in_channels,
self.nf,
self.embed_dim,
self.ch_mult,
self.n_blocks,
self.resolution,
2023-03-03 06:02:00 +00:00
self.attn_resolutions,
)
if self.quantizer_type == "nearest":
2023-03-03 06:02:00 +00:00
self.beta = beta # 0.25
self.quantize = VectorQuantizer(
self.codebook_size, self.embed_dim, self.beta
)
elif self.quantizer_type == "gumbel":
self.gumbel_num_hiddens = emb_dim
self.straight_through = gumbel_straight_through
self.kl_weight = gumbel_kl_weight
self.quantize = GumbelQuantizer(
self.codebook_size,
self.embed_dim,
self.gumbel_num_hiddens,
self.straight_through,
2023-03-03 06:02:00 +00:00
self.kl_weight,
)
self.generator = Generator(
self.nf,
self.embed_dim,
self.ch_mult,
self.n_blocks,
self.resolution,
2023-03-03 06:02:00 +00:00
self.attn_resolutions,
)
if model_path is not None:
2023-03-03 06:02:00 +00:00
chkpt = torch.load(model_path, map_location="cpu")
if "params_ema" in chkpt:
self.load_state_dict(
torch.load(model_path, map_location="cpu")["params_ema"]
)
logger.info(f"vqgan is loaded from: {model_path} [params_ema]")
elif "params" in chkpt:
self.load_state_dict(
torch.load(model_path, map_location="cpu")["params"]
)
logger.info(f"vqgan is loaded from: {model_path} [params]")
else:
2023-03-03 06:02:00 +00:00
raise ValueError(f"Wrong params!")
def forward(self, x):
x = self.encoder(x)
quant, codebook_loss, quant_stats = self.quantize(x)
x = self.generator(quant)
return x, codebook_loss, quant_stats
# patch based discriminator
@ARCH_REGISTRY.register()
class VQGANDiscriminator(nn.Module):
def __init__(self, nc=3, ndf=64, n_layers=4, model_path=None):
super().__init__()
2023-03-03 06:02:00 +00:00
layers = [
nn.Conv2d(nc, ndf, kernel_size=4, stride=2, padding=1),
nn.LeakyReLU(0.2, True),
]
ndf_mult = 1
ndf_mult_prev = 1
for n in range(1, n_layers): # gradually increase the number of filters
ndf_mult_prev = ndf_mult
2023-03-03 06:02:00 +00:00
ndf_mult = min(2**n, 8)
layers += [
2023-03-03 06:02:00 +00:00
nn.Conv2d(
ndf * ndf_mult_prev,
ndf * ndf_mult,
kernel_size=4,
stride=2,
padding=1,
bias=False,
),
nn.BatchNorm2d(ndf * ndf_mult),
2023-03-03 06:02:00 +00:00
nn.LeakyReLU(0.2, True),
]
ndf_mult_prev = ndf_mult
2023-03-03 06:02:00 +00:00
ndf_mult = min(2**n_layers, 8)
layers += [
2023-03-03 06:02:00 +00:00
nn.Conv2d(
ndf * ndf_mult_prev,
ndf * ndf_mult,
kernel_size=4,
stride=1,
padding=1,
bias=False,
),
nn.BatchNorm2d(ndf * ndf_mult),
2023-03-03 06:02:00 +00:00
nn.LeakyReLU(0.2, True),
]
layers += [
2023-03-03 06:02:00 +00:00
nn.Conv2d(ndf * ndf_mult, 1, kernel_size=4, stride=1, padding=1)
] # output 1 channel prediction map
self.main = nn.Sequential(*layers)
if model_path is not None:
2023-03-03 06:02:00 +00:00
chkpt = torch.load(model_path, map_location="cpu")
if "params_d" in chkpt:
self.load_state_dict(
torch.load(model_path, map_location="cpu")["params_d"]
)
elif "params" in chkpt:
self.load_state_dict(
torch.load(model_path, map_location="cpu")["params"]
)
else:
2023-03-03 06:02:00 +00:00
raise ValueError(f"Wrong params!")
def forward(self, x):
return self.main(x)