xuman-Amy / preprocessingLinks
数据预处理之缺失值处理,特征选择
☆21Updated 6 years ago
Alternatives and similar repositories for preprocessing
Users that are interested in preprocessing are comparing it to the libraries listed below
Sorting:
- 常用的特征选择方法☆68Updated 3 years ago
- 基于遗传算法的特征选择☆127Updated 5 years ago
- 机器学习的特征工程,包括特 征抽取、特征预处理、特征选择、特征降维。☆25Updated 6 years ago
- Bayesian Optimization and Grid Search for xgboost/lightgbm☆74Updated 4 months ago
- 第三届“融360”天机智能算法挑战赛中“拒绝推断”赛题--复赛第四名的代码分享☆11Updated 5 years ago
- Oversampling method based on relative density☆13Updated 4 years ago
- Oversampling for imbalanced learning based on k-means and SMOTE☆127Updated 4 years ago
- 机器学习集成模型之Stacking各类模型及工具源码☆117Updated 4 years ago
- TensorFlow Probability;Time series model☆127Updated 3 years ago
- 使用sklearn做特征工程☆172Updated 6 years ago
- 马上AI全球挑战赛-违约用户风 险预测 top2-solution☆18Updated 7 years ago
- 国内首个迁移学习赛题 中国平安前海征信“好信杯”迁移学习大数据算法大赛 FInSight团队作品(算法方案排名第三)☆87Updated 6 years ago
- 异常检测☆159Updated 6 years ago
- 2019年CCF智能信用评分大赛个人源码库。包含XGboost模型调参,特征筛选,训练等方案。同时包含stacking模型融合方案☆27Updated 5 years ago
- 2020 招商银行FinTech 数据赛道 rank10☆115Updated 5 years ago
- The code of the AAAI-19 paper "AFS: An Attention-based mechanism for Supervised Feature Selection".☆46Updated 6 years ago
- Transfer Learning JDA and TrAdaboost☆65Updated 7 years ago
- Time Series Prediction, Stateful LSTM; 时间序列预测,洗发水销量/股票走势预测,有状态循环神经网络☆58Updated 7 years ago
- Cost-Sensitive Learning / ReSampling / Weighting / Thresholding / BorderlineSMOTE / AdaCost / etc.☆107Updated 4 years ago
- Affinity Propagation Clustering with DTW distance on temporal sequence classification☆20Updated 6 years ago
- 集成学习Stacking方法详解☆75Updated 5 years ago
- 使用遗传算法结合决策树做特征选择/Using genetic algorithm for feature selection with decision tree☆25Updated 7 years ago
- 在sklearn下,几种常用的特征选择方法☆40Updated 9 years ago
- 天池-印象盐城-汽车销量预测大赛☆36Updated 7 years ago
- ☆101Updated 6 years ago
- 数据特征工程、各种机器学习回归模型、回归数据预处理☆44Updated 5 years ago
- 当样本分布发生变化时,交叉验证无法准确评估模型在测试集上的效果,这时候需要其他构造验证集的方法来应对。☆51Updated 5 years ago
- 通过科研人员论文项目等数据,训练识别导师/学生的分类器。代码包括特征选择基础、网格搜索确定特征选择方法参数、不平衡数据的处理(oversampling、undersampling)和pu-learning方法在此问题上的应用☆30Updated 6 years ago
- 分类类别不平衡,解决办法:采样(SMOTE和算法集成技术等)、阈值移动、调整代价或权重,附带信用卡诈骗案例☆21Updated 5 years ago
- 包括决策树和随机森林进行离职人员预测,Xgboost和lightGBM的应用☆289Updated 5 years ago