Project-Practice

practice [baseline]

[NLP]-结合实体关系信息的用户喜好预测，使用MKR结构，info

[CV]-简易表情分类，info

[CV]-Kaggle木薯叶病分类，info

[NLP]-简易self attention文本分类，info

[NLP]-简易textCNN文本分类，info

[NLP]-疫情期间情绪识别，info

[NLP]-文本分类capsule net， info

[NLP]-文本中风险事件识别(only recommendation)，[info](./[NLP]-文本中风险事件识别(only recommendation))

[ML]-地震预测，时序预测，info

[ML]-Kaggle GPS 定位修正，回归预测，info

[ML]-多目标视频推荐，分类，info

[ML]-工业蒸汽量预测，多模型集成回归预测问题，易过拟合, 线上0.1140, 198/9319 info

[ML]-天猫复购预测，分类, 线上0.6830249, 158/5349 info

[ML]-天池O2O，二分类，购买概率预测 1089/20224 info

[ML]-云安全监测，分类，437/2654 info

[ML]-Kaggle Predict Future Sales，时序回归, 线上0.89272, 941 / 10,789 info

[ML]-Kaggle House Prices - Advanced Regression Techniques，回归, 线上0.11976, 439 / 6,835 info

[NLP]-Kaggle CommonLit Readablity - Regression, rmse 0.461没有提交了最后实验得到的几个stack模型 [info](./[NLP]-kaggle-CommonLit Readablity)