2022-11-29 12:22:53 -07:00
|
|
|
# The content of this file comes from the ldm/models/autoencoder.py file of the compvis/stable-diffusion repo
|
|
|
|
# The VQModel & VQModelInterface were subsequently removed from ldm/models/autoencoder.py when we moved to the stability-ai/stablediffusion repo
|
|
|
|
# As the LDSR upscaler relies on VQModel & VQModelInterface, the hijack aims to put them back into the ldm.models.autoencoder
|
2023-05-09 23:25:25 -06:00
|
|
|
import numpy as np
|
2022-11-29 10:38:16 -07:00
|
|
|
import torch
|
|
|
|
import pytorch_lightning as pl
|
|
|
|
import torch.nn.functional as F
|
|
|
|
from contextlib import contextmanager
|
2023-05-09 23:25:25 -06:00
|
|
|
|
|
|
|
from torch.optim.lr_scheduler import LambdaLR
|
|
|
|
|
|
|
|
from ldm.modules.ema import LitEma
|
2023-05-30 03:36:55 -06:00
|
|
|
from vqvae_quantize import VectorQuantizer2 as VectorQuantizer
|
2022-11-29 10:38:16 -07:00
|
|
|
from ldm.modules.diffusionmodules.model import Encoder, Decoder
|
|
|
|
from ldm.util import instantiate_from_config
|
|
|
|
|
|
|
|
import ldm.models.autoencoder
|
2023-05-09 23:25:25 -06:00
|
|
|
from packaging import version
|
2022-11-29 10:38:16 -07:00
|
|
|
|
|
|
|
class VQModel(pl.LightningModule):
|
|
|
|
def __init__(self,
|
|
|
|
ddconfig,
|
|
|
|
lossconfig,
|
|
|
|
n_embed,
|
|
|
|
embed_dim,
|
|
|
|
ckpt_path=None,
|
2023-05-10 02:19:16 -06:00
|
|
|
ignore_keys=None,
|
2022-11-29 10:38:16 -07:00
|
|
|
image_key="image",
|
|
|
|
colorize_nlabels=None,
|
|
|
|
monitor=None,
|
|
|
|
batch_resize_range=None,
|
|
|
|
scheduler_config=None,
|
|
|
|
lr_g_factor=1.0,
|
|
|
|
remap=None,
|
|
|
|
sane_index_shape=False, # tell vector quantizer to return indices as bhw
|
|
|
|
use_ema=False
|
|
|
|
):
|
|
|
|
super().__init__()
|
|
|
|
self.embed_dim = embed_dim
|
|
|
|
self.n_embed = n_embed
|
|
|
|
self.image_key = image_key
|
|
|
|
self.encoder = Encoder(**ddconfig)
|
|
|
|
self.decoder = Decoder(**ddconfig)
|
|
|
|
self.loss = instantiate_from_config(lossconfig)
|
|
|
|
self.quantize = VectorQuantizer(n_embed, embed_dim, beta=0.25,
|
|
|
|
remap=remap,
|
|
|
|
sane_index_shape=sane_index_shape)
|
|
|
|
self.quant_conv = torch.nn.Conv2d(ddconfig["z_channels"], embed_dim, 1)
|
|
|
|
self.post_quant_conv = torch.nn.Conv2d(embed_dim, ddconfig["z_channels"], 1)
|
|
|
|
if colorize_nlabels is not None:
|
|
|
|
assert type(colorize_nlabels)==int
|
|
|
|
self.register_buffer("colorize", torch.randn(3, colorize_nlabels, 1, 1))
|
|
|
|
if monitor is not None:
|
|
|
|
self.monitor = monitor
|
|
|
|
self.batch_resize_range = batch_resize_range
|
|
|
|
if self.batch_resize_range is not None:
|
|
|
|
print(f"{self.__class__.__name__}: Using per-batch resizing in range {batch_resize_range}.")
|
|
|
|
|
|
|
|
self.use_ema = use_ema
|
|
|
|
if self.use_ema:
|
|
|
|
self.model_ema = LitEma(self)
|
|
|
|
print(f"Keeping EMAs of {len(list(self.model_ema.buffers()))}.")
|
|
|
|
|
|
|
|
if ckpt_path is not None:
|
2023-05-10 02:19:16 -06:00
|
|
|
self.init_from_ckpt(ckpt_path, ignore_keys=ignore_keys or [])
|
2022-11-29 10:38:16 -07:00
|
|
|
self.scheduler_config = scheduler_config
|
|
|
|
self.lr_g_factor = lr_g_factor
|
|
|
|
|
|
|
|
@contextmanager
|
|
|
|
def ema_scope(self, context=None):
|
|
|
|
if self.use_ema:
|
|
|
|
self.model_ema.store(self.parameters())
|
|
|
|
self.model_ema.copy_to(self)
|
|
|
|
if context is not None:
|
|
|
|
print(f"{context}: Switched to EMA weights")
|
|
|
|
try:
|
|
|
|
yield None
|
|
|
|
finally:
|
|
|
|
if self.use_ema:
|
|
|
|
self.model_ema.restore(self.parameters())
|
|
|
|
if context is not None:
|
|
|
|
print(f"{context}: Restored training weights")
|
|
|
|
|
2023-05-10 02:19:16 -06:00
|
|
|
def init_from_ckpt(self, path, ignore_keys=None):
|
2022-11-29 10:38:16 -07:00
|
|
|
sd = torch.load(path, map_location="cpu")["state_dict"]
|
|
|
|
keys = list(sd.keys())
|
|
|
|
for k in keys:
|
2023-05-10 02:19:16 -06:00
|
|
|
for ik in ignore_keys or []:
|
2022-11-29 10:38:16 -07:00
|
|
|
if k.startswith(ik):
|
|
|
|
print("Deleting key {} from state_dict.".format(k))
|
|
|
|
del sd[k]
|
|
|
|
missing, unexpected = self.load_state_dict(sd, strict=False)
|
|
|
|
print(f"Restored from {path} with {len(missing)} missing and {len(unexpected)} unexpected keys")
|
2023-06-02 05:58:10 -06:00
|
|
|
if missing:
|
2022-11-29 10:38:16 -07:00
|
|
|
print(f"Missing Keys: {missing}")
|
2023-06-02 05:58:10 -06:00
|
|
|
if unexpected:
|
2022-11-29 10:38:16 -07:00
|
|
|
print(f"Unexpected Keys: {unexpected}")
|
|
|
|
|
|
|
|
def on_train_batch_end(self, *args, **kwargs):
|
|
|
|
if self.use_ema:
|
|
|
|
self.model_ema(self)
|
|
|
|
|
|
|
|
def encode(self, x):
|
|
|
|
h = self.encoder(x)
|
|
|
|
h = self.quant_conv(h)
|
|
|
|
quant, emb_loss, info = self.quantize(h)
|
|
|
|
return quant, emb_loss, info
|
|
|
|
|
|
|
|
def encode_to_prequant(self, x):
|
|
|
|
h = self.encoder(x)
|
|
|
|
h = self.quant_conv(h)
|
|
|
|
return h
|
|
|
|
|
|
|
|
def decode(self, quant):
|
|
|
|
quant = self.post_quant_conv(quant)
|
|
|
|
dec = self.decoder(quant)
|
|
|
|
return dec
|
|
|
|
|
|
|
|
def decode_code(self, code_b):
|
|
|
|
quant_b = self.quantize.embed_code(code_b)
|
|
|
|
dec = self.decode(quant_b)
|
|
|
|
return dec
|
|
|
|
|
|
|
|
def forward(self, input, return_pred_indices=False):
|
|
|
|
quant, diff, (_,_,ind) = self.encode(input)
|
|
|
|
dec = self.decode(quant)
|
|
|
|
if return_pred_indices:
|
|
|
|
return dec, diff, ind
|
|
|
|
return dec, diff
|
|
|
|
|
|
|
|
def get_input(self, batch, k):
|
|
|
|
x = batch[k]
|
|
|
|
if len(x.shape) == 3:
|
|
|
|
x = x[..., None]
|
|
|
|
x = x.permute(0, 3, 1, 2).to(memory_format=torch.contiguous_format).float()
|
|
|
|
if self.batch_resize_range is not None:
|
|
|
|
lower_size = self.batch_resize_range[0]
|
|
|
|
upper_size = self.batch_resize_range[1]
|
|
|
|
if self.global_step <= 4:
|
|
|
|
# do the first few batches with max size to avoid later oom
|
|
|
|
new_resize = upper_size
|
|
|
|
else:
|
|
|
|
new_resize = np.random.choice(np.arange(lower_size, upper_size+16, 16))
|
|
|
|
if new_resize != x.shape[2]:
|
|
|
|
x = F.interpolate(x, size=new_resize, mode="bicubic")
|
|
|
|
x = x.detach()
|
|
|
|
return x
|
|
|
|
|
|
|
|
def training_step(self, batch, batch_idx, optimizer_idx):
|
|
|
|
# https://github.com/pytorch/pytorch/issues/37142
|
|
|
|
# try not to fool the heuristics
|
|
|
|
x = self.get_input(batch, self.image_key)
|
|
|
|
xrec, qloss, ind = self(x, return_pred_indices=True)
|
|
|
|
|
|
|
|
if optimizer_idx == 0:
|
|
|
|
# autoencode
|
|
|
|
aeloss, log_dict_ae = self.loss(qloss, x, xrec, optimizer_idx, self.global_step,
|
|
|
|
last_layer=self.get_last_layer(), split="train",
|
|
|
|
predicted_indices=ind)
|
|
|
|
|
|
|
|
self.log_dict(log_dict_ae, prog_bar=False, logger=True, on_step=True, on_epoch=True)
|
|
|
|
return aeloss
|
|
|
|
|
|
|
|
if optimizer_idx == 1:
|
|
|
|
# discriminator
|
|
|
|
discloss, log_dict_disc = self.loss(qloss, x, xrec, optimizer_idx, self.global_step,
|
|
|
|
last_layer=self.get_last_layer(), split="train")
|
|
|
|
self.log_dict(log_dict_disc, prog_bar=False, logger=True, on_step=True, on_epoch=True)
|
|
|
|
return discloss
|
|
|
|
|
|
|
|
def validation_step(self, batch, batch_idx):
|
|
|
|
log_dict = self._validation_step(batch, batch_idx)
|
|
|
|
with self.ema_scope():
|
2023-05-09 22:52:45 -06:00
|
|
|
self._validation_step(batch, batch_idx, suffix="_ema")
|
2022-11-29 10:38:16 -07:00
|
|
|
return log_dict
|
|
|
|
|
|
|
|
def _validation_step(self, batch, batch_idx, suffix=""):
|
|
|
|
x = self.get_input(batch, self.image_key)
|
|
|
|
xrec, qloss, ind = self(x, return_pred_indices=True)
|
|
|
|
aeloss, log_dict_ae = self.loss(qloss, x, xrec, 0,
|
|
|
|
self.global_step,
|
|
|
|
last_layer=self.get_last_layer(),
|
|
|
|
split="val"+suffix,
|
|
|
|
predicted_indices=ind
|
|
|
|
)
|
|
|
|
|
|
|
|
discloss, log_dict_disc = self.loss(qloss, x, xrec, 1,
|
|
|
|
self.global_step,
|
|
|
|
last_layer=self.get_last_layer(),
|
|
|
|
split="val"+suffix,
|
|
|
|
predicted_indices=ind
|
|
|
|
)
|
|
|
|
rec_loss = log_dict_ae[f"val{suffix}/rec_loss"]
|
|
|
|
self.log(f"val{suffix}/rec_loss", rec_loss,
|
|
|
|
prog_bar=True, logger=True, on_step=False, on_epoch=True, sync_dist=True)
|
|
|
|
self.log(f"val{suffix}/aeloss", aeloss,
|
|
|
|
prog_bar=True, logger=True, on_step=False, on_epoch=True, sync_dist=True)
|
|
|
|
if version.parse(pl.__version__) >= version.parse('1.4.0'):
|
|
|
|
del log_dict_ae[f"val{suffix}/rec_loss"]
|
|
|
|
self.log_dict(log_dict_ae)
|
|
|
|
self.log_dict(log_dict_disc)
|
|
|
|
return self.log_dict
|
|
|
|
|
|
|
|
def configure_optimizers(self):
|
|
|
|
lr_d = self.learning_rate
|
|
|
|
lr_g = self.lr_g_factor*self.learning_rate
|
|
|
|
print("lr_d", lr_d)
|
|
|
|
print("lr_g", lr_g)
|
|
|
|
opt_ae = torch.optim.Adam(list(self.encoder.parameters())+
|
|
|
|
list(self.decoder.parameters())+
|
|
|
|
list(self.quantize.parameters())+
|
|
|
|
list(self.quant_conv.parameters())+
|
|
|
|
list(self.post_quant_conv.parameters()),
|
|
|
|
lr=lr_g, betas=(0.5, 0.9))
|
|
|
|
opt_disc = torch.optim.Adam(self.loss.discriminator.parameters(),
|
|
|
|
lr=lr_d, betas=(0.5, 0.9))
|
|
|
|
|
|
|
|
if self.scheduler_config is not None:
|
|
|
|
scheduler = instantiate_from_config(self.scheduler_config)
|
|
|
|
|
|
|
|
print("Setting up LambdaLR scheduler...")
|
|
|
|
scheduler = [
|
|
|
|
{
|
|
|
|
'scheduler': LambdaLR(opt_ae, lr_lambda=scheduler.schedule),
|
|
|
|
'interval': 'step',
|
|
|
|
'frequency': 1
|
|
|
|
},
|
|
|
|
{
|
|
|
|
'scheduler': LambdaLR(opt_disc, lr_lambda=scheduler.schedule),
|
|
|
|
'interval': 'step',
|
|
|
|
'frequency': 1
|
|
|
|
},
|
|
|
|
]
|
|
|
|
return [opt_ae, opt_disc], scheduler
|
|
|
|
return [opt_ae, opt_disc], []
|
|
|
|
|
|
|
|
def get_last_layer(self):
|
|
|
|
return self.decoder.conv_out.weight
|
|
|
|
|
|
|
|
def log_images(self, batch, only_inputs=False, plot_ema=False, **kwargs):
|
2023-05-10 02:55:09 -06:00
|
|
|
log = {}
|
2022-11-29 10:38:16 -07:00
|
|
|
x = self.get_input(batch, self.image_key)
|
|
|
|
x = x.to(self.device)
|
|
|
|
if only_inputs:
|
|
|
|
log["inputs"] = x
|
|
|
|
return log
|
|
|
|
xrec, _ = self(x)
|
|
|
|
if x.shape[1] > 3:
|
|
|
|
# colorize with random projection
|
|
|
|
assert xrec.shape[1] > 3
|
|
|
|
x = self.to_rgb(x)
|
|
|
|
xrec = self.to_rgb(xrec)
|
|
|
|
log["inputs"] = x
|
|
|
|
log["reconstructions"] = xrec
|
|
|
|
if plot_ema:
|
|
|
|
with self.ema_scope():
|
|
|
|
xrec_ema, _ = self(x)
|
2023-05-09 23:25:25 -06:00
|
|
|
if x.shape[1] > 3:
|
|
|
|
xrec_ema = self.to_rgb(xrec_ema)
|
2022-11-29 10:38:16 -07:00
|
|
|
log["reconstructions_ema"] = xrec_ema
|
|
|
|
return log
|
|
|
|
|
|
|
|
def to_rgb(self, x):
|
|
|
|
assert self.image_key == "segmentation"
|
|
|
|
if not hasattr(self, "colorize"):
|
|
|
|
self.register_buffer("colorize", torch.randn(3, x.shape[1], 1, 1).to(x))
|
|
|
|
x = F.conv2d(x, weight=self.colorize)
|
|
|
|
x = 2.*(x-x.min())/(x.max()-x.min()) - 1.
|
|
|
|
return x
|
|
|
|
|
|
|
|
|
|
|
|
class VQModelInterface(VQModel):
|
|
|
|
def __init__(self, embed_dim, *args, **kwargs):
|
2023-05-10 02:19:16 -06:00
|
|
|
super().__init__(*args, embed_dim=embed_dim, **kwargs)
|
2022-11-29 10:38:16 -07:00
|
|
|
self.embed_dim = embed_dim
|
|
|
|
|
|
|
|
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
|
|
|
|
|
2023-05-10 02:05:02 -06:00
|
|
|
ldm.models.autoencoder.VQModel = VQModel
|
|
|
|
ldm.models.autoencoder.VQModelInterface = VQModelInterface
|