Spaces:
Sleeping
Sleeping
#--- START OF MODIFIED FILE app_fluxContext_Ltx/ltx_worker_upscaler.py --- | |
# ltx_worker_upscaler.py | |
# Worker para fazer upscale de latentes de vídeo para alta resolução. | |
# Este arquivo é parte do projeto Euia-AducSdr e está sob a licença AGPL v3. | |
# Copyright (C) 4 de Agosto de 2025 Carlos Rodrigues dos Santos | |
import torch | |
import gc | |
import os | |
import yaml | |
import numpy as np | |
import imageio | |
from pathlib import Path | |
import huggingface_hub | |
from PIL import Image # <--- IMPORTAÇÃO ADICIONADA AQUI | |
from inference import create_ltx_video_pipeline | |
from ltx_video.models.autoencoders.latent_upsampler import LatentUpsampler | |
from ltx_video.models.autoencoders.vae_encode import vae_encode, vae_decode | |
class LtxUpscaler: | |
def __init__(self, device_id='cuda:0'): | |
print(f"WORKER CÂMERA-UPSCALER: Inicializando para {device_id}...") | |
self.device = torch.device(device_id if torch.cuda.is_available() else 'cpu') | |
self.model_dtype = torch.bfloat16 | |
config_file_path = "configs/ltxv-13b-0.9.8-dev.yaml" | |
with open(config_file_path, "r") as file: | |
self.config = yaml.safe_load(file) | |
LTX_REPO = "Lightricks/LTX-Video" | |
models_dir = "downloaded_models_gradio" | |
Path(models_dir).mkdir(parents=True, exist_ok=True) | |
print(f"WORKER CÂMERA-UPSCALER ({self.device}): Carregando VAE na CPU...") | |
model_actual_path = huggingface_hub.hf_hub_download( | |
repo_id=LTX_REPO, filename=self.config["checkpoint_path"], | |
local_dir=models_dir, local_dir_use_symlinks=False | |
) | |
temp_pipeline = create_ltx_video_pipeline( | |
ckpt_path=model_actual_path, precision=self.config["precision"], | |
text_encoder_model_name_or_path=self.config["text_encoder_model_name_or_path"], | |
sampler=self.config["sampler"], device='cpu' | |
) | |
self.vae = temp_pipeline.vae.to(self.model_dtype) | |
del temp_pipeline | |
gc.collect() | |
print(f"WORKER CÂMERA-UPSCALER ({self.device}): Carregando Latent Upsampler na CPU...") | |
upscaler_path = huggingface_hub.hf_hub_download( | |
repo_id=LTX_REPO, filename=self.config["spatial_upscaler_model_path"], | |
local_dir=models_dir, local_dir_use_symlinks=False | |
) | |
self.latent_upsampler = LatentUpsampler.from_pretrained(upscaler_path).to(self.model_dtype) | |
self.latent_upsampler.to('cpu') | |
self.vae.to('cpu') | |
print(f"WORKER CÂMERA-UPSCALER ({self.device}): Pronto (na CPU).") | |
def to_gpu(self): | |
if self.latent_upsampler and self.vae and torch.cuda.is_available(): | |
print(f"WORKER CÂMERA-UPSCALER: Movendo modelos para {self.device}...") | |
self.latent_upsampler.to(self.device) | |
self.vae.to(self.device) | |
def to_cpu(self): | |
if self.latent_upsampler and self.vae: | |
print(f"WORKER CÂMERA-UPSCALER: Descarregando modelos da GPU {self.device}...") | |
self.latent_upsampler.to('cpu') | |
self.vae.to('cpu') | |
gc.collect() | |
if torch.cuda.is_available(): | |
torch.cuda.empty_cache() | |
def upscale_latents_to_video(self, latent_path: str, output_path: str, video_fps: int): | |
print(f"UPSCALER ({self.device}): Processando latentes de {os.path.basename(latent_path)}") | |
latents = torch.load(latent_path).to(self.device, dtype=self.model_dtype) | |
upsampled_latents = self.latent_upsampler(latents) | |
decode_timestep = torch.tensor([0.0] * upsampled_latents.shape[0], device=self.device) | |
upsampled_video_tensor = vae_decode( | |
upsampled_latents, self.vae, is_video=True, timestep=decode_timestep | |
) | |
upsampled_video_tensor = (upsampled_video_tensor.clamp(-1, 1) + 1) / 2.0 | |
video_np_high_res = (upsampled_video_tensor[0].permute(1, 2, 3, 0).cpu().float().numpy() * 255).astype(np.uint8) | |
with imageio.get_writer(output_path, fps=video_fps, codec='libx264', quality=8) as writer: | |
for frame in video_np_high_res: | |
writer.append_data(frame) | |
print(f"UPSCALER ({self.device}): Arquivo de vídeo salvo em {os.path.basename(output_path)}") | |
return output_path | |
def decode_single_latent_frame(self, latent_frame_tensor: torch.Tensor) -> Image.Image: | |
"""Decodifica um único frame latente para uma imagem PIL para o Gemini.""" | |
latent_frame_tensor = latent_frame_tensor.to(self.device, dtype=self.model_dtype) | |
decode_timestep = torch.tensor([0.0] * latent_frame_tensor.shape[0], device=self.device) | |
decoded_tensor = vae_decode( | |
latent_frame_tensor, self.vae, is_video=True, timestep=decode_timestep | |
) | |
decoded_tensor = (decoded_tensor.clamp(-1, 1) + 1) / 2.0 | |
numpy_image = (decoded_tensor[0].permute(2, 3, 1, 0).squeeze().cpu().float().numpy() * 255).astype(np.uint8) | |
return Image.fromarray(numpy_image) | |
#--- END OF MODIFIED FILE app_fluxContext_Ltx/ltx_worker_upscaler.py --- |