Content-Length: 235642 | pFad | http://github.com/isLouisHsu/tianchi_gaic_track1_share

9C GitHub - isLouisHsu/tianchi_gaic_track1_share: 2021.3.15 周周星分享
Skip to content

isLouisHsu/tianchi_gaic_track1_share

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 

Repository files navigation

大家好,我是来自华东理工大学的“单键”团队的单键丶,很荣幸能拿到本周的周周星。

方案分享如下

  • 模型结构选用nezha-cn-base,尝试使用nezha-cn-large,线下过拟合严重;
  • 预训练MLM模型,参数重新初始化,试错调整默认mask参数,lr=5e-5, epochs=10,mlm损失最终0.17左右;
  • 微调多标签分类模型,即pooled_output接MLP进行各标签二分类,采用二分类交叉熵作为损失函数;
  • 随机划分五折交叉验证,线下单模型logloss=0.995左右,线上0.908左右,差距较大,五折融合后达到0.910左右;
  • 线上线下数据差距较大,可以尝试不同的划分方法(当然也有可能是数据量不足的问题);
  • 调参!调参!调参!数据增强、对抗训练;
  • EDA无效、添加tfidf特征无效。

最后,祝大家LB暴涨!

About

2021.3.15 周周星分享

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published








ApplySandwichStrip

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier!      Saves Data!


--- a PPN by Garber Painting Akron. With Image Size Reduction included!

Fetched URL: http://github.com/isLouisHsu/tianchi_gaic_track1_share

Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy