Skip to content

Releases: PaulYangSz/MyJavaLogisticRegression

Add AdaBoost

04 Apr 05:13
Compare
Choose a tag to compare

10 Base LR model with alpha=1.0, iterTimes=100.
Result is better than LR(0.001, 100), but weaker than LR(0.001, 1000).

Complete the first kaggle-titanic predict, with Hxy feature engineering and gradient decent optimal method.

30 Mar 11:47
Compare
Choose a tag to compare

v1.0版只是用来表明自己对于一些理论有了一次实际的尝试,其实远不到发布版本的时候,不过用以记录下罢了。

接下来可能要去暂时学别的模型了,回头有机会还会回来再看,再挖掘数据的特征,对模型的最优解求解进行别的方法尝试,并比较不同的参数。