1.target encoding
先整理一下鏈接,之後會看。
簡介入門:
https://zhuanlan.zhihu.com/p/40231966
一個各種category 變量編碼的庫:
https://github.com/scikit-learn-contrib/categorical-encoding
說明文檔:
http://contrib.scikit-learn.org/categorical-encoding/targetencoder.html
一個英文的關於它的說明
http://www.saedsayad.com/encoding.htm
kaggle上實現target encoding的一個kernel:
https://www.kaggle.com/ogrellier/python-target-encoding-for-categorical-features
how lightgbm deals with categorical features:
https://www.kaggle.com/c/home-credit-default-risk/discussion/58950
2.count encoding
離散變量與較少取值的連續變量可用