github
/
minimind-v
mirror of https://github.com/jingyaogong/minimind-v.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Compare commits

base: github:a9e02b0ff61a551db3c0ea3aeaabdae7f352a161
Branches Tags
github:master
github:website
github:v2
github:v1
..
compare: github:7c4e935bde56ff126b772f174b6556f04f9a3be6
Branches Tags
github:master
github:website
github:v2
github:v1

2 Commits
a9e02b0ff6 ... 7c4e935bde

Author SHA1 Message Date
dynorth 7c4e935bde
Merge 2a65e6a675 into 2b94dcb3d0 2026-05-14 11:27:57 +08:00
jingyaogong 2b94dcb3d0 [update] num_workers=2 2026-05-12 09:54:46 +08:00
2 changed files with 2 additions and 2 deletions
Show Stats Expand all files Collapse all files

2
trainer/train_pretrain_vlm.py
View File

@ -95,7 +95,7 @@ if __name__ == "__main__":
parser.add_argument("--learning_rate", type=float, default=4e-4, help="初始学习率") parser.add_argument("--learning_rate", type=float, default=4e-4, help="初始学习率")
parser.add_argument("--device", type=str, default="cuda:0" if torch.cuda.is_available() else "cpu", help="训练设备") parser.add_argument("--device", type=str, default="cuda:0" if torch.cuda.is_available() else "cpu", help="训练设备")
parser.add_argument("--dtype", type=str, default="bfloat16", help="混合精度类型") parser.add_argument("--dtype", type=str, default="bfloat16", help="混合精度类型")
parser.add_argument("--num_workers", type=int, default=8, help="数据加载线程数") parser.add_argument("--num_workers", type=int, default=2, help="数据加载线程数")
parser.add_argument("--accumulation_steps", type=int, default=1, help="梯度累积步数") parser.add_argument("--accumulation_steps", type=int, default=1, help="梯度累积步数")
parser.add_argument("--grad_clip", type=float, default=1.0, help="梯度裁剪阈值") parser.add_argument("--grad_clip", type=float, default=1.0, help="梯度裁剪阈值")
parser.add_argument("--log_interval", type=int, default=100, help="日志打印间隔") parser.add_argument("--log_interval", type=int, default=100, help="日志打印间隔")

2
trainer/train_sft_vlm.py
View File

@ -95,7 +95,7 @@ if __name__ == "__main__":
parser.add_argument("--learning_rate", type=float, default=5e-6, help="初始学习率") parser.add_argument("--learning_rate", type=float, default=5e-6, help="初始学习率")
parser.add_argument("--device", type=str, default="cuda:0" if torch.cuda.is_available() else "cpu", help="训练设备") parser.add_argument("--device", type=str, default="cuda:0" if torch.cuda.is_available() else "cpu", help="训练设备")
parser.add_argument("--dtype", type=str, default="bfloat16", help="混合精度类型") parser.add_argument("--dtype", type=str, default="bfloat16", help="混合精度类型")
parser.add_argument("--num_workers", type=int, default=8, help="数据加载线程数") parser.add_argument("--num_workers", type=int, default=2, help="数据加载线程数")
parser.add_argument("--accumulation_steps", type=int, default=1, help="梯度累积步数") parser.add_argument("--accumulation_steps", type=int, default=1, help="梯度累积步数")
parser.add_argument("--grad_clip", type=float, default=1.0, help="梯度裁剪阈值") parser.add_argument("--grad_clip", type=float, default=1.0, help="梯度裁剪阈值")
parser.add_argument("--log_interval", type=int, default=100, help="日志打印间隔") parser.add_argument("--log_interval", type=int, default=100, help="日志打印间隔")