GPT-SoVITS
GPT-SoVITS copied to clipboard
RVC-Boss
GPT训练：NameError: name 'trainer' is not defined
之前试过两次，训练都挺正常的，然后中途删了项目，重新git clone了一个最新版的，到了GPT训练这一步，总是报错。在s1_train.py的140行左右
try:
# 使用正则表达式匹配文件名中的数字部分，并按数字大小进行排序
newest_ckpt_name = get_newest_ckpt(os.listdir(ckpt_dir))
ckpt_path = ckpt_dir / newest_ckpt_name
except Exception:
ckpt_path = None
print("ckpt_path:", ckpt_path)
trainer.fit(model, data_module, ckpt_path=ckpt_path)
尝试过在trainer.fit()方法之前打印trainer，model，data_module，都能够打印出来，说明应该是定义了的，但是这个trainer接下来就总是报错，问各位大佬，是哪里的低级操作错误 我没有意识到吗
"/Users/mao/miniconda3/envs/GPTSoVits/bin/python" GPT_SoVITS/s1_train.py --config_file "/Users/mao/Downloads/pyproject/GPT-SoVITS/TEMP/tmp_s1.yaml" Seed set to 1234 Using 16bit Automatic Mixed Precision (AMP) /Users/mao/miniconda3/envs/GPTSoVits/lib/python3.9/site-packages/torch/cuda/amp/grad_scaler.py:126: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available. Disabling. warnings.warn( GPU available: True (mps), used: True TPU available: False, using: 0 TPU cores IPU available: False, using: 0 IPUs HPU available: False, using: 0 HPUs <All keys matched successfully> ckpt_path: None semantic_data_len: 15 phoneme_data_len: 15 item_name semantic_audio 0 guodegang-full.MP3_0000005760_0000283200.wav 536 486 41 185 656 965 965 965 774 565 623 298... 1 guodegang-full.MP3_0000594240_0000851520.wav 208 103 216 232 305 604 142 462 551 369 572 26... 2 guodegang-full.MP3_0001112640_0001374080.wav 208 365 232 257 366 641 123 964 854 5 500 5 96... 3 guodegang-full.MP3_0001669120_0001929920.wav 921 519 939 840 208 884 406 237 42 989 680 626... 4 guodegang-full.MP3_0002185280_0002443840.wav 208 318 656 797 328 998 998 40 312 917 776 531... 5 guodegang-full.MP3_0002793920_0003050560.wav 208 965 318 656 965 837 758 687 933 422 606 54... 6 guodegang-full.MP3_0003352000_0003617280.wav 208 318 947 698 424 1018 662 846 846 846 329 6... 7 guodegang-full.MP3_0003933120_0004135040.wav 76 2 834 205 549 820 900 8 17 918 42 672 942 2... 8 guodegang-full.MP3_0000292480_0000594240.wav 837 208 837 160 787 1016 402 1017 795 555 808 ... 9 guodegang-full.MP3_0000851520_0001100160.wav 208 103 656 774 837 434 183 123 949 320 469 93... 10 guodegang-full.MP3_0001374080_0001669120.wav 214 844 14 851 439 794 777 356 777 400 569 420... 11 guodegang-full.MP3_0001929920_0002185280.wav 3 942 45 559 253 229 629 827 216 437 765 841 8... 12 guodegang-full.MP3_0002449280_0002792320.wav 208 837 837 612 582 921 657 106 409 590 341 92... 13 guodegang-full.MP3_0003050560_0003342400.wav 208 103 208 1005 837 1005 103 656 194 982 654 ... 14 guodegang-full.MP3_0003617280_0003933120.wav 1005 248 132 592 592 592 785 211 392 19 723 75... dataset.len(): 90
| Name | Type | Params
0 | model | Text2SemanticDecoder | 77.5 M
77.5 M Trainable params
0 Non-trainable params
77.5 M Total params
309.975 Total estimated model params size (MB)
Traceback (most recent call last):
File "/Users/mao/Downloads/pyproject/GPT-SoVITS/GPT_SoVITS/s1_train.py", line 170, in
