# 1. make_data.sh  (生成特征数据)
# 2. stat_freq.sh (统计特征频次，过滤低频特征)
# 3. pipline/make_train_sample.sh(生成训练数据，使用过滤低频的特征文件)
# 4. pipline/train_xgb_model.sh ([训练模型](https://git.yishihui.com/algorithm/recommend-model/src/feature/jch))
# 5. pipline/xgb_model_eval.sh ([评估模型](https://git.yishihui.com/algorithm/recommend-model/src/feature/jch))
# 6. pipline/eval_qq.sh ([生成qq](https://git.yishihui.com/algorithm/recommend-model/src/feature/jch))