Merge branch 'dev' into optimizer-expand-and-refactor

library/train_util.py

@@ -1407,6 +1407,12 @@ def add_training_arguments(parser: argparse.ArgumentParser, support_dreambooth:
   parser.add_argument("--train_batch_size", type=int, default=1, help="batch size for training / 学習時のバッチサイズ")
   parser.add_argument("--max_token_length", type=int, default=None, choices=[None, 150, 225],
                       help="max token length of text encoder (default for 75, 150 or 225) / text encoderのトークンの最大長（未指定で75、150または225が指定可）")
+  # parser.add_argument("--use_8bit_adam", action="store_true",
+  #                     help="use 8bit Adam optimizer (requires bitsandbytes) / 8bit Adamオプティマイザを使う（bitsandbytesのインストールが必要）")
+  # parser.add_argument("--use_lion_optimizer", action="store_true",
+  #                     help="use Lion optimizer (requires lion-pytorch) / Lionオプティマイザを使う（ lion-pytorch のインストールが必要）")
+  # parser.add_argument("--use_dadaptation_optimizer", action="store_true",
+  #                     help="use dadaptation optimizer (requires dadaptation) / dadaptaionオプティマイザを使う（ dadaptation のインストールが必要）")
   parser.add_argument("--mem_eff_attn", action="store_true",
                       help="use memory efficient attention for CrossAttention / CrossAttentionに省メモリ版attentionを使う")
   parser.add_argument("--xformers", action="store_true",
@@ -1514,7 +1520,7 @@ def add_sd_saving_arguments(parser: argparse.ArgumentParser):
 
 # region utils
 
-# "Optimizer to use: AdamW, AdamW8bit, Lion, SGDNesterov, SGDNesterov8bit"
+# "Optimizer to use: AdamW, AdamW8bit, Lion, SGDNesterov, SGDNesterov8bit, dadaption"
 
 def get_optimizer(args, trainable_params):
   # Prepare optimizer/学習に必要なクラスを準備する
@@ -1557,6 +1563,17 @@ def get_optimizer(args, trainable_params):
     optimizer_class = torch.optim.SGD
     optimizer = optimizer_class(trainable_params, lr=lr, momentum=momentum, weight_decay=weight_decay, nesterov=True)
 
+  elif optimizer_type == "dadaptation".lower():
+    try:
+      import dadaptation
+    except ImportError:
+      raise ImportError("No dadaptation / dadaptation がインストールされていないようです")
+    print(f"use dadaptation optimizer")
+    optimizer_class = dadaptation.DAdaptAdam
+    if args.learning_rate <= 0.1:
+      print('learning rate is too low. If using dadaptaion, set learning rate around 1.0.')
+      print('recommend option: lr=1.0')
+    optimizer = optimizer_class(trainable_params, lr=lr)
   else:
     print(f"use AdamW optimizer | betas: {betas}, Weight Decay: {weight_decay}")
     optimizer_class = torch.optim.AdamW

train_network.py

@@ -37,6 +37,9 @@ def generate_step_logs(args: argparse.Namespace, current_loss, avr_loss, lr_sche
     logs["lr/textencoder"] = lr_scheduler.get_last_lr()[0]
     logs["lr/unet"] = lr_scheduler.get_last_lr()[-1]          # may be same to textencoder
 
+  if args.use_dadaptation_optimizer: # tracking d*lr value of unet.
+    logs["lr/d*lr"] = lr_scheduler.optimizers[-1].param_groups[0]['d']*lr_scheduler.optimizers[-1].param_groups[0]['lr']
+
   return logs