|
|
|
|
|
import torch |
|
import os |
|
from tqdm import tqdm |
|
|
|
from toolkit import train_tools |
|
from toolkit.prompt_utils import PromptEmbeds |
|
from toolkit.stable_diffusion_model import StableDiffusion |
|
|
|
|
|
def mu_tilde(model, xt, x0, timestep): |
|
"mu_tilde(x_t, x_0) DDPM paper eq. 7" |
|
prev_timestep = timestep - model.scheduler.config.num_train_timesteps // model.scheduler.num_inference_steps |
|
alpha_prod_t_prev = model.scheduler.alphas_cumprod[ |
|
prev_timestep] if prev_timestep >= 0 else model.scheduler.final_alpha_cumprod |
|
alpha_t = model.scheduler.alphas[timestep] |
|
beta_t = 1 - alpha_t |
|
alpha_bar = model.scheduler.alphas_cumprod[timestep] |
|
return ((alpha_prod_t_prev ** 0.5 * beta_t) / (1 - alpha_bar)) * x0 + ( |
|
(alpha_t ** 0.5 * (1 - alpha_prod_t_prev)) / (1 - alpha_bar)) * xt |
|
|
|
|
|
def sample_xts_from_x0(sd: StableDiffusion, sample: torch.Tensor, num_inference_steps=50): |
|
""" |
|
Samples from P(x_1:T|x_0) |
|
""" |
|
|
|
alpha_bar = sd.noise_scheduler.alphas_cumprod |
|
sqrt_one_minus_alpha_bar = (1 - alpha_bar) ** 0.5 |
|
alphas = sd.noise_scheduler.alphas |
|
betas = 1 - alphas |
|
|
|
|
|
|
|
|
|
|
|
variance_noise_shape = list(sample.shape) |
|
variance_noise_shape[0] = num_inference_steps |
|
|
|
timesteps = sd.noise_scheduler.timesteps.to(sd.device) |
|
t_to_idx = {int(v): k for k, v in enumerate(timesteps)} |
|
xts = torch.zeros(variance_noise_shape).to(sample.device, dtype=torch.float16) |
|
for t in reversed(timesteps): |
|
idx = t_to_idx[int(t)] |
|
xts[idx] = sample * (alpha_bar[t] ** 0.5) + torch.randn_like(sample, dtype=torch.float16) * sqrt_one_minus_alpha_bar[t] |
|
xts = torch.cat([xts, sample], dim=0) |
|
|
|
return xts |
|
|
|
|
|
def encode_text(model, prompts): |
|
text_input = model.tokenizer( |
|
prompts, |
|
padding="max_length", |
|
max_length=model.tokenizer.model_max_length, |
|
truncation=True, |
|
return_tensors="pt", |
|
) |
|
with torch.no_grad(): |
|
text_encoding = model.text_encoder(text_input.input_ids.to(model.device))[0] |
|
return text_encoding |
|
|
|
|
|
def forward_step(sd: StableDiffusion, model_output, timestep, sample): |
|
next_timestep = min( |
|
sd.noise_scheduler.config['num_train_timesteps'] - 2, |
|
timestep + sd.noise_scheduler.config['num_train_timesteps'] // sd.noise_scheduler.num_inference_steps |
|
) |
|
|
|
|
|
alpha_prod_t = sd.noise_scheduler.alphas_cumprod[timestep] |
|
|
|
|
|
beta_prod_t = 1 - alpha_prod_t |
|
|
|
|
|
|
|
pred_original_sample = (sample - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5) |
|
|
|
|
|
next_sample = sd.noise_scheduler.add_noise( |
|
pred_original_sample, |
|
model_output, |
|
torch.LongTensor([next_timestep])) |
|
return next_sample |
|
|
|
|
|
def get_variance(sd: StableDiffusion, timestep): |
|
prev_timestep = timestep - sd.noise_scheduler.config['num_train_timesteps'] // sd.noise_scheduler.num_inference_steps |
|
alpha_prod_t = sd.noise_scheduler.alphas_cumprod[timestep] |
|
alpha_prod_t_prev = sd.noise_scheduler.alphas_cumprod[ |
|
prev_timestep] if prev_timestep >= 0 else sd.noise_scheduler.final_alpha_cumprod |
|
beta_prod_t = 1 - alpha_prod_t |
|
beta_prod_t_prev = 1 - alpha_prod_t_prev |
|
variance = (beta_prod_t_prev / beta_prod_t) * (1 - alpha_prod_t / alpha_prod_t_prev) |
|
return variance |
|
|
|
|
|
def get_time_ids_from_latents(sd: StableDiffusion, latents: torch.Tensor): |
|
VAE_SCALE_FACTOR = 2 ** (len(sd.vae.config['block_out_channels']) - 1) |
|
if sd.is_xl: |
|
bs, ch, h, w = list(latents.shape) |
|
|
|
height = h * VAE_SCALE_FACTOR |
|
width = w * VAE_SCALE_FACTOR |
|
|
|
dtype = latents.dtype |
|
|
|
target_size = (height, width) |
|
original_size = (height, width) |
|
crops_coords_top_left = (0, 0) |
|
add_time_ids = list(original_size + crops_coords_top_left + target_size) |
|
add_time_ids = torch.tensor([add_time_ids]) |
|
add_time_ids = add_time_ids.to(latents.device, dtype=dtype) |
|
|
|
batch_time_ids = torch.cat( |
|
[add_time_ids for _ in range(bs)] |
|
) |
|
return batch_time_ids |
|
else: |
|
return None |
|
|
|
|
|
def inversion_forward_process( |
|
sd: StableDiffusion, |
|
sample: torch.Tensor, |
|
conditional_embeddings: PromptEmbeds, |
|
unconditional_embeddings: PromptEmbeds, |
|
etas=None, |
|
prog_bar=False, |
|
cfg_scale=3.5, |
|
num_inference_steps=50, eps=None |
|
): |
|
current_num_timesteps = len(sd.noise_scheduler.timesteps) |
|
sd.noise_scheduler.set_timesteps(num_inference_steps, device=sd.device) |
|
|
|
timesteps = sd.noise_scheduler.timesteps.to(sd.device) |
|
|
|
|
|
|
|
|
|
|
|
|
|
variance_noise_shape = list(sample.shape) |
|
variance_noise_shape[0] = num_inference_steps |
|
if etas is None or (type(etas) in [int, float] and etas == 0): |
|
eta_is_zero = True |
|
zs = None |
|
else: |
|
eta_is_zero = False |
|
if type(etas) in [int, float]: etas = [etas] * sd.noise_scheduler.num_inference_steps |
|
xts = sample_xts_from_x0(sd, sample, num_inference_steps=num_inference_steps) |
|
alpha_bar = sd.noise_scheduler.alphas_cumprod |
|
zs = torch.zeros(size=variance_noise_shape, device=sd.device, dtype=torch.float16) |
|
|
|
t_to_idx = {int(v): k for k, v in enumerate(timesteps)} |
|
noisy_sample = sample |
|
op = tqdm(reversed(timesteps), desc="Inverting...") if prog_bar else reversed(timesteps) |
|
|
|
for timestep in op: |
|
idx = t_to_idx[int(timestep)] |
|
|
|
if not eta_is_zero: |
|
noisy_sample = xts[idx][None] |
|
|
|
added_cond_kwargs = {} |
|
|
|
with torch.no_grad(): |
|
text_embeddings = train_tools.concat_prompt_embeddings( |
|
unconditional_embeddings, |
|
conditional_embeddings, |
|
1, |
|
) |
|
if sd.is_xl: |
|
add_time_ids = get_time_ids_from_latents(sd, noisy_sample) |
|
|
|
add_time_ids = torch.cat( |
|
[add_time_ids] * 2, dim=0 |
|
) |
|
|
|
added_cond_kwargs = { |
|
"text_embeds": text_embeddings.pooled_embeds, |
|
"time_ids": add_time_ids, |
|
} |
|
|
|
|
|
latent_model_input = torch.cat( |
|
[noisy_sample] * 2, dim=0 |
|
) |
|
|
|
noise_pred = sd.unet( |
|
latent_model_input, |
|
timestep, |
|
encoder_hidden_states=text_embeddings.text_embeds, |
|
added_cond_kwargs=added_cond_kwargs, |
|
).sample |
|
|
|
noise_pred_uncond, noise_pred_text = noise_pred.chunk(2) |
|
|
|
|
|
|
|
|
|
noise_pred = noise_pred_uncond + cfg_scale * (noise_pred_text - noise_pred_uncond) |
|
|
|
if eta_is_zero: |
|
|
|
noisy_sample = forward_step(sd, noise_pred, timestep, noisy_sample) |
|
xts = None |
|
|
|
else: |
|
xtm1 = xts[idx + 1][None] |
|
|
|
pred_original_sample = (noisy_sample - (1 - alpha_bar[timestep]) ** 0.5 * noise_pred) / alpha_bar[ |
|
timestep] ** 0.5 |
|
|
|
|
|
prev_timestep = timestep - sd.noise_scheduler.config[ |
|
'num_train_timesteps'] // sd.noise_scheduler.num_inference_steps |
|
alpha_prod_t_prev = sd.noise_scheduler.alphas_cumprod[ |
|
prev_timestep] if prev_timestep >= 0 else sd.noise_scheduler.final_alpha_cumprod |
|
|
|
variance = get_variance(sd, timestep) |
|
pred_sample_direction = (1 - alpha_prod_t_prev - etas[idx] * variance) ** (0.5) * noise_pred |
|
|
|
mu_xt = alpha_prod_t_prev ** (0.5) * pred_original_sample + pred_sample_direction |
|
|
|
z = (xtm1 - mu_xt) / (etas[idx] * variance ** 0.5) |
|
zs[idx] = z |
|
|
|
|
|
xtm1 = mu_xt + (etas[idx] * variance ** 0.5) * z |
|
xts[idx + 1] = xtm1 |
|
|
|
if not zs is None: |
|
zs[-1] = torch.zeros_like(zs[-1]) |
|
|
|
|
|
sd.noise_scheduler.set_timesteps(current_num_timesteps, device=sd.device) |
|
|
|
return noisy_sample, zs, xts |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def reverse_step(model, model_output, timestep, sample, eta=0, variance_noise=None): |
|
|
|
prev_timestep = timestep - model.scheduler.config.num_train_timesteps // model.scheduler.num_inference_steps |
|
|
|
alpha_prod_t = model.scheduler.alphas_cumprod[timestep] |
|
alpha_prod_t_prev = model.scheduler.alphas_cumprod[ |
|
prev_timestep] if prev_timestep >= 0 else model.scheduler.final_alpha_cumprod |
|
beta_prod_t = 1 - alpha_prod_t |
|
|
|
|
|
pred_original_sample = (sample - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5) |
|
|
|
|
|
|
|
variance = get_variance(model, timestep) |
|
std_dev_t = eta * variance ** (0.5) |
|
|
|
model_output_direction = model_output |
|
|
|
|
|
pred_sample_direction = (1 - alpha_prod_t_prev - eta * variance) ** (0.5) * model_output_direction |
|
|
|
prev_sample = alpha_prod_t_prev ** (0.5) * pred_original_sample + pred_sample_direction |
|
|
|
if eta > 0: |
|
if variance_noise is None: |
|
variance_noise = torch.randn(model_output.shape, device=model.device, dtype=torch.float16) |
|
sigma_z = eta * variance ** (0.5) * variance_noise |
|
prev_sample = prev_sample + sigma_z |
|
|
|
return prev_sample |
|
|
|
|
|
def inversion_reverse_process( |
|
model, |
|
xT, |
|
etas=0, |
|
prompts="", |
|
cfg_scales=None, |
|
prog_bar=False, |
|
zs=None, |
|
controller=None, |
|
asyrp=False): |
|
batch_size = len(prompts) |
|
|
|
cfg_scales_tensor = torch.Tensor(cfg_scales).view(-1, 1, 1, 1).to(model.device, dtype=torch.float16) |
|
|
|
text_embeddings = encode_text(model, prompts) |
|
uncond_embedding = encode_text(model, [""] * batch_size) |
|
|
|
if etas is None: etas = 0 |
|
if type(etas) in [int, float]: etas = [etas] * model.scheduler.num_inference_steps |
|
assert len(etas) == model.scheduler.num_inference_steps |
|
timesteps = model.scheduler.timesteps.to(model.device) |
|
|
|
xt = xT.expand(batch_size, -1, -1, -1) |
|
op = tqdm(timesteps[-zs.shape[0]:]) if prog_bar else timesteps[-zs.shape[0]:] |
|
|
|
t_to_idx = {int(v): k for k, v in enumerate(timesteps[-zs.shape[0]:])} |
|
|
|
for t in op: |
|
idx = t_to_idx[int(t)] |
|
|
|
with torch.no_grad(): |
|
uncond_out = model.unet.forward(xt, timestep=t, |
|
encoder_hidden_states=uncond_embedding) |
|
|
|
|
|
if prompts: |
|
with torch.no_grad(): |
|
cond_out = model.unet.forward(xt, timestep=t, |
|
encoder_hidden_states=text_embeddings) |
|
|
|
z = zs[idx] if not zs is None else None |
|
z = z.expand(batch_size, -1, -1, -1) |
|
if prompts: |
|
|
|
noise_pred = uncond_out.sample + cfg_scales_tensor * (cond_out.sample - uncond_out.sample) |
|
else: |
|
noise_pred = uncond_out.sample |
|
|
|
xt = reverse_step(model, noise_pred, t, xt, eta=etas[idx], variance_noise=z) |
|
if controller is not None: |
|
xt = controller.step_callback(xt) |
|
return xt, zs |
|
|