All Projects → yongyehuang → Dc Hi_guides

yongyehuang / Dc Hi_guides

[Data Castle 算法竞赛] 精品旅行服务成单预测 final rank 11

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Dc Hi guides

Mlbox
MLBox is a powerful Automated Machine Learning python library.
Stars: ✭ 1,199 (+1344.58%)
Mutual labels:  xgboost, lightgbm, stacking
Tencent2017 Final Rank28 code
2017第一届腾讯社交广告高校算法大赛Rank28_code
Stars: ✭ 85 (+2.41%)
Mutual labels:  data-mining, xgboost, lightgbm
My Data Competition Experience
本人多次机器学习与大数据竞赛Top5的经验总结,满满的干货,拿好不谢
Stars: ✭ 271 (+226.51%)
Mutual labels:  xgboost, lightgbm
Open Solution Home Credit
Open solution to the Home Credit Default Risk challenge 🏡
Stars: ✭ 397 (+378.31%)
Mutual labels:  xgboost, lightgbm
Rong360
用户贷款风险预测
Stars: ✭ 489 (+489.16%)
Mutual labels:  data-mining, xgboost
diabetes use case
Sample use case for Xavier AI in Healthcare conference: https://www.xavierhealth.org/ai-summit-day2/
Stars: ✭ 22 (-73.49%)
Mutual labels:  data-mining, xgboost
HousePrice
住房月租金预测大数据赛TOP1
Stars: ✭ 17 (-79.52%)
Mutual labels:  xgboost, lightgbm
Ai competitions
AI比赛相关信息汇总
Stars: ✭ 443 (+433.73%)
Mutual labels:  xgboost, lightgbm
ai-deployment
关注AI模型上线、模型部署
Stars: ✭ 149 (+79.52%)
Mutual labels:  xgboost, lightgbm
Hyperparameter hunter
Easy hyperparameter optimization and automatic result saving across machine learning algorithms and libraries
Stars: ✭ 648 (+680.72%)
Mutual labels:  xgboost, lightgbm
Awesome Gradient Boosting Papers
A curated list of gradient boosting research papers with implementations.
Stars: ✭ 704 (+748.19%)
Mutual labels:  xgboost, lightgbm
HyperGBM
A full pipeline AutoML tool for tabular data
Stars: ✭ 172 (+107.23%)
Mutual labels:  xgboost, lightgbm
HumanOrRobot
a solution for competition of kaggle `Human or Robot`
Stars: ✭ 16 (-80.72%)
Mutual labels:  xgboost, lightgbm
Leaves
pure Go implementation of prediction part for GBRT (Gradient Boosting Regression Trees) models from popular frameworks
Stars: ✭ 261 (+214.46%)
Mutual labels:  xgboost, lightgbm
Arch-Data-Science
Archlinux PKGBUILDs for Data Science, Machine Learning, Deep Learning, NLP and Computer Vision
Stars: ✭ 92 (+10.84%)
Mutual labels:  xgboost, lightgbm
Mli Resources
H2O.ai Machine Learning Interpretability Resources
Stars: ✭ 428 (+415.66%)
Mutual labels:  data-mining, xgboost
Mljar Supervised
Automated Machine Learning Pipeline with Feature Engineering and Hyper-Parameters Tuning 🚀
Stars: ✭ 961 (+1057.83%)
Mutual labels:  xgboost, lightgbm
xgboost-lightgbm-hyperparameter-tuning
Bayesian Optimization and Grid Search for xgboost/lightgbm
Stars: ✭ 40 (-51.81%)
Mutual labels:  xgboost, lightgbm
Openscoring
REST web service for the true real-time scoring (<1 ms) of Scikit-Learn, R and Apache Spark models
Stars: ✭ 536 (+545.78%)
Mutual labels:  xgboost, lightgbm
Text Classification Benchmark
文本分类基准测试
Stars: ✭ 18 (-78.31%)
Mutual labels:  xgboost, lightgbm

HI GUIDES 精品旅行服务成单预测

  • final rank: 11

比赛说明:精品旅行服务成单预测

  • 竞赛背景: 第二届智慧中国杯首发 皇包车(HI GUIDES)是一个为中国出境游用户提供全球中文包车游服务的平台。

  • 比赛目标:我们提供了5万多名用户在旅游app中的浏览行为记录,其中有些用户在浏览之后完成了订单,且享受了精品旅游服务,而有些用户则没有下单。参赛者需要分析用户的个人信息和浏览行为,从而预测用户是否会在短期内购买精品旅游服务。预测用户是否会在短期内购买精品旅游服务

文件结构

|- hi_guide
|  |- data        # 比赛提供的原始数据
|  |  |- test      |  |  |- trainingset       |  |- features      # 特征提取函数
|  |  |- action.py     # 行为特征
|  |  |- comment.py     # 评论特征
|  |  |- history.py     # 历史订单特征
|  |  |- profile.py     # 用户信息特征
|  |  |- train_data.csv   # 保存提取的训练特征  
|  |  |- test_data.csv   # 保存提取的测试集特征 
|  |- log        # 模型训练日志
|  |- result       # 模型预测结果
|  |- model       # 保存训练好的模型和特征重要度分析文件
|  |- data_helper.py  # 执行特征提取的代码
|  |- my_utils.py   # 工具函数库,主要用到其中的 xgb 特征重要度分析函数
|  |- m1_xgb.py   # xgb 模型
|  |- m2_lgb.py   # lgb 模型
|  |- m3_cgb.py   # catboost 模型
|  |- stacking.py   # stacking 模型融合
|  |- get_no_used_features.py   # 获取 xgb 和 lgb 中的特征重要度

使用方式

# run the single model
python -u m1_xgb.py
# run the stacking model
python -u stacking.py
# 注意事项
# train_data, test_data = load_feat(re_get=True, feature_path=feature_path)  # 如果没有修改特征,设置re_get=False,就会直接导入之前保存好的特征。
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].