Merge pull request #107 from kohya-ss/dev

merge dev to main

fine_tune.py

@@ -200,6 +200,8 @@ def train(args):
   # epoch数を計算する
   num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
   num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
+  if (args.save_n_epoch_ratio is not None) and (args.save_n_epoch_ratio > 0):
+    args.save_every_n_epochs = math.floor(num_train_epochs / args.save_n_epoch_ratio) or 1
 
   # 学習する
   total_batch_size = args.train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps

library/train_util.py

@@ -12,6 +12,7 @@ import math
 import os
 import random
 import hashlib
+from io import BytesIO
 
 from tqdm import tqdm
 import torch
@@ -25,6 +26,7 @@ from PIL import Image
 import cv2
 from einops import rearrange
 from torch import einsum
+import safetensors.torch
 
 import library.model_util as model_util
 
@@ -85,6 +87,7 @@ class BaseDataset(torch.utils.data.Dataset):
     self.enable_bucket = False
     self.min_bucket_reso = None
     self.max_bucket_reso = None
+    self.bucket_info = None
 
     self.tokenizer_max_length = self.tokenizer.model_max_length if max_token_length is None else max_token_length + 2
 
@@ -217,11 +220,17 @@ class BaseDataset(torch.utils.data.Dataset):
         self.buckets[bucket_index].append(image_info.image_key)
 
     if self.enable_bucket:
+      self.bucket_info = {"buckets": {}}
       print("number of images (including repeats) / 各bucketの画像枚数（繰り返し回数を含む）")
       for i, (reso, img_keys) in enumerate(zip(bucket_resos, self.buckets)):
+        self.bucket_info["buckets"][i] = {"resolution": reso, "count": len(img_keys)}
         print(f"bucket {i}: resolution {reso}, count: {len(img_keys)}")
+
       img_ar_errors = np.array(img_ar_errors)
-      print(f"mean ar error (without repeats): {np.mean(np.abs(img_ar_errors))}")
+      mean_img_ar_error = np.mean(np.abs(img_ar_errors))
+      self.bucket_info["mean_img_ar_error"] = mean_img_ar_error
+      print(f"mean ar error (without repeats): {mean_img_ar_error}")
+      
 
     # 参照用indexを作る
     self.buckets_indices: list(BucketBatchIndex) = []
@@ -790,6 +799,49 @@ def calculate_sha256(filename):
   return hash_sha256.hexdigest()
 
 
+def precalculate_safetensors_hashes(tensors, metadata):
+  """Precalculate the model hashes needed by sd-webui-additional-networks to
+  save time on indexing the model later."""
+
+  # Because writing user metadata to the file can change the result of
+  # sd_models.model_hash(), only retain the training metadata for purposes of
+  # calculating the hash, as they are meant to be immutable
+  metadata = {k: v for k, v in metadata.items() if k.startswith("ss_")}
+
+  bytes = safetensors.torch.save(tensors, metadata)
+  b = BytesIO(bytes)
+
+  model_hash = addnet_hash_safetensors(b)
+  legacy_hash = addnet_hash_legacy(b)
+  return model_hash, legacy_hash
+
+
+def addnet_hash_legacy(b):
+  """Old model hash used by sd-webui-additional-networks for .safetensors format files"""
+  m = hashlib.sha256()
+
+  b.seek(0x100000)
+  m.update(b.read(0x10000))
+  return m.hexdigest()[0:8]
+
+
+def addnet_hash_safetensors(b):
+  """New model hash used by sd-webui-additional-networks for .safetensors format files"""
+  hash_sha256 = hashlib.sha256()
+  blksize = 1024 * 1024
+
+  b.seek(0)
+  header = b.read(8)
+  n = int.from_bytes(header, "little")
+
+  offset = n + 8
+  b.seek(offset)
+  for chunk in iter(lambda: b.read(blksize), b""):
+    hash_sha256.update(chunk)
+
+  return hash_sha256.hexdigest()
+
+
 # flash attention forwards and backwards
 
 # https://arxiv.org/abs/2205.14135
@@ -1057,6 +1109,8 @@ def add_training_arguments(parser: argparse.ArgumentParser, support_dreambooth:
                       choices=[None, "float", "fp16", "bf16"], help="precision in saving / 保存時に精度を変更して保存する")
   parser.add_argument("--save_every_n_epochs", type=int, default=None,
                       help="save checkpoint every N epochs / 学習中のモデルを指定エポックごとに保存する")
+  parser.add_argument("--save_n_epoch_ratio", type=int, default=None,
+                      help="save checkpoint N epoch ratio (for example 5 means save at least 5 files total) / 学習中のモデルを指定のエポック割合で保存する（たとえば5を指定すると最低5個のファイルが保存される）")
   parser.add_argument("--save_last_n_epochs", type=int, default=None, help="save last N checkpoints / 最大Nエポック保存する")
   parser.add_argument("--save_last_n_epochs_state", type=int, default=None,
                       help="save last N checkpoints of state (overrides the value of --save_last_n_epochs)/ 最大Nエポックstateを保存する(--save_last_n_epochsの指定を上書きします)")

networks/lora.py

@@ -7,6 +7,8 @@ import math
 import os
 import torch
 
+from library import train_util
+
 
 class LoRAModule(torch.nn.Module):
   """
@@ -31,7 +33,7 @@ class LoRAModule(torch.nn.Module):
       self.lora_up = torch.nn.Linear(lora_dim, out_dim, bias=False)
 
     if type(alpha) == torch.Tensor:
-      alpha = alpha.detach().numpy()
+      alpha = alpha.detach().float().numpy()                              # without casting, bf16 causes error
     alpha = lora_dim if alpha is None or alpha == 0 else alpha
     self.scale = alpha / self.lora_dim
     self.register_buffer('alpha', torch.tensor(alpha))                    # 定数として扱える
@@ -221,6 +223,14 @@ class LoRANetwork(torch.nn.Module):
 
     if os.path.splitext(file)[1] == '.safetensors':
       from safetensors.torch import save_file
+
+      # Precalculate model hashes to save time on indexing
+      if metadata is None:
+        metadata = {}
+      model_hash, legacy_hash = train_util.precalculate_safetensors_hashes(state_dict, metadata)
+      metadata["sshs_model_hash"] = model_hash
+      metadata["sshs_legacy_hash"] = legacy_hash
+
       save_file(state_dict, file, metadata)
     else:
       torch.save(state_dict, file)

train_db.py

@@ -176,6 +176,8 @@ def train(args):
   # epoch数を計算する
   num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
   num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
+  if (args.save_n_epoch_ratio is not None) and (args.save_n_epoch_ratio > 0):
+    args.save_every_n_epochs = math.floor(num_train_epochs / args.save_n_epoch_ratio) or 1
 
   # 学習する
   total_batch_size = args.train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps

train_network.py

@@ -212,6 +212,8 @@ def train(args):
   # epoch数を計算する
   num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
   num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
+  if (args.save_n_epoch_ratio is not None) and (args.save_n_epoch_ratio > 0):
+    args.save_every_n_epochs = math.floor(num_train_epochs / args.save_n_epoch_ratio) or 1
 
   # 学習する
   total_batch_size = args.train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps
@@ -264,6 +266,7 @@ def train(args):
       "ss_keep_tokens": args.keep_tokens,
       "ss_dataset_dirs": json.dumps(train_dataset.dataset_dirs_info),
       "ss_reg_dataset_dirs": json.dumps(train_dataset.reg_dataset_dirs_info),
+      "ss_bucket_info": json.dumps(train_dataset.bucket_info),
       "ss_training_comment": args.training_comment        # will not be updated after training
   }
 
@@ -437,8 +440,8 @@ if __name__ == '__main__':
   train_util.add_training_arguments(parser, True)
 
   parser.add_argument("--no_metadata", action='store_true', help="do not save metadata in output model / メタデータを出力先モデルに保存しない")
-  parser.add_argument("--save_model_as", type=str, default="pt", choices=[None, "ckpt", "pt", "safetensors"],
-                      help="format to save the model (default is .pt) / モデル保存時の形式（デフォルトはpt）")
+  parser.add_argument("--save_model_as", type=str, default="safetensors", choices=[None, "ckpt", "pt", "safetensors"],
+                      help="format to save the model (default is .safetensors) / モデル保存時の形式（デフォルトはsafetensors）")
 
   parser.add_argument("--unet_lr", type=float, default=None, help="learning rate for U-Net / U-Netの学習率")
   parser.add_argument("--text_encoder_lr", type=float, default=None, help="learning rate for Text Encoder / Text Encoderの学習率")