Akegarasu / Akegarasu/lora-scripts
奇怪了，flux一训练就报错，设置也没问题啊，是4090D的，用的是flux-dev.sft模型
- Lenguaje dominante
- Python
- Estrellas
- 6.1k
- Forks
- 699
- Métricas de merge de PR
- Sin PR fusionados en 30 d
Descripción
Traceback (most recent call last):
File "E:\【LoRA训练】SD-Trainer\lora-scripts-v1.5.1\lora-scripts-v1.5.1\scripts\dev\flux_train_network.py", line 574, in
trainer.train(args)
File "E:\【LoRA训练】SD-Trainer\lora-scripts-v1.5.1\lora-scripts-v1.5.1\scripts\dev\train_network.py", line 388, in train
model_version, text_encoder, vae, unet = self.load_target_model(args, weight_dtype, accelerator)
File "E:\【LoRA训练】SD-Trainer\lora-scripts-v1.5.1\lora-scripts-v1.5.1\scripts\dev\flux_train_network.py", line 82, in load_target_model
self.is_schnell, model = flux_utils.load_flow_model(
File "E:\【LoRA训练】SD-Trainer\lora-scripts-v1.5.1\lora-scripts-v1.5.1\scripts\dev\library\flux_utils.py", line 136, in load_flow_model
info = model.load_state_dict(sd, strict=False, assign=True)
TypeError: Module.load_state_dict() got an unexpected keyword argument 'assign'
14:19:55-071299 ERROR Training failed / 训练失败
Guía de contribución
No hay ninguna guía de contribución indexada para este repositorio
Evaluación
Este issue todavía no se ha evaluado.