| from diffusers import DiffusionPipeline | |
| import os | |
| class FluxPipeline(DiffusionPipeline): | |
| def __init__( | |
| self, | |
| vae, | |
| text_encoder, | |
| text_encoder_2, | |
| tokenizer, | |
| tokenizer_2, | |
| transformer, | |
| scheduler, | |
| **kwargs | |
| ): | |
| super().__init__() | |
| self.vae = vae | |
| self.text_encoder = text_encoder | |
| self.text_encoder_2 = text_encoder_2 | |
| self.tokenizer = tokenizer | |
| self.tokenizer_2 = tokenizer_2 | |
| self.transformer = transformer | |
| self.scheduler = scheduler | |
| # сюда можно добавить доп. обработку kwargs | |
| for k, v in kwargs.items(): | |
| setattr(self, k, v) | |
| def load_attn_procs(self, path: str): | |
| if not os.path.exists(path): | |
| raise FileNotFoundError(f"LoRA file not found: {path}") | |
| print(f"[FluxPipeline] Loading LoRA from {path}") | |
| self.load_lora_weights(path) | |