jiahui
/
text-classification-long
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
排版识别标题级别和正文
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
2 Commits
1 Branch
0 Tags
548 KiB
 
 
Branch: master
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'master'
${ noResults }
text-classification-long/README.md

1.1 KiB
Raw Permalink Blame History

训练脚本

bash bash run_train_2.sh

[INFO|trainer.py:2144] 2025-05-13 16:26:32,249 >> ***** Running training *****
[INFO|trainer.py:2145] 2025-05-13 16:26:32,249 >>   Num examples = 2,699
[INFO|trainer.py:2146] 2025-05-13 16:26:32,249 >>   Num Epochs = 5
[INFO|trainer.py:2147] 2025-05-13 16:26:32,249 >>   Instantaneous batch size per device = 1
[INFO|trainer.py:2150] 2025-05-13 16:26:32,249 >>   Total train batch size (w. parallel, distributed & accumulation) = 1
[INFO|trainer.py:2151] 2025-05-13 16:26:32,249 >>   Gradient Accumulation steps = 1
[INFO|trainer.py:2152] 2025-05-13 16:26:32,249 >>   Total optimization steps = 13,495
[INFO|trainer.py:2153] 2025-05-13 16:26:32,250 >>   Number of trainable parameters = 105,023,236
{'loss': 1.6225, 'grad_norm': 0.6827912926673889, 'learning_rate': 1.925898480918859e-05, 'epoch': 0.19}                                                     
  4%|████▏                                                                                                             | 500/13495 [03:34<1:33:16,  2.32it/s]

测试效果

python 批量测试结果.py