black

Author: AlexanderDokuchaev

examples/openvino/stable-diffusion/train_text_to_image_qat.py

@@ -59,6 +59,7 @@
     DIFFUSION_MODEL_VAE_ENCODER_SUBFOLDER,
 )
 
+
 requests.packages.urllib3.disable_warnings(InsecureRequestWarning)
 
 random.seed(42)
@@ -993,9 +994,7 @@ def collate_fn(examples):
         args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch
         overrode_max_train_steps = True
 
-    optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
-        optimizer, train_dataloader, lr_scheduler
-    )
+    optimizer, train_dataloader, lr_scheduler = accelerator.prepare(optimizer, train_dataloader, lr_scheduler)
 
     weight_dtype = torch.float32
     if args.mixed_precision == "fp16":