Fix DDP issues and Support DDP for all training scripts (#448)

* Fix DDP bugs

* Fix DDP bugs for finetune and db

* refactor model loader

* fix DDP network

* try to fix DDP network in train unet only

* remove unuse DDP import

* refactor DDP transform

* refactor DDP transform

* fix sample images bugs

* change DDP tranform location

* add autocast to train_db

* support DDP in XTI

* Clear DDP import
This commit is contained in:
Isotr0py
2023-05-03 09:37:47 +08:00
committed by GitHub
parent a7485e4d9e
commit e1143caf38
7 changed files with 58 additions and 37 deletions

View File

@@ -23,7 +23,7 @@ def interrogate(args):
print(f"loading SD model: {args.sd_model}")
args.pretrained_model_name_or_path = args.sd_model
args.vae = None
text_encoder, vae, unet, _ = train_util.load_target_model(args,weights_dtype, DEVICE)
text_encoder, vae, unet, _ = train_util._load_target_model(args,weights_dtype, DEVICE)
print(f"loading LoRA: {args.model}")
network, weights_sd = lora.create_network_from_weights(1.0, args.model, vae, text_encoder, unet)