美文网首页
Underfitting and overfitting

Underfitting and overfitting

作者: BillLeee | 来源:发表于2017-04-25 14:23 被阅读0次

    一、underfitting 和 overfitting 中,error的变化情况

    IMG_20170425_135852.jpg

    Underfitting:
    training error will high, and cross validation error also will be high.
    Overfitting:
    training error will be low,but cross validation will be low.

    二、在regularization中,lambda对error的影响

    IMG_20170425_144508.jpg

    随着lambda 的增加,training error在增加,整个过程由overfitting->fitting->underfitting,所以cross validation由大->小->大。
    实验中,需要不断调整lambda的值(一般从0逐渐增加,step可以为0.01,或者0.1等),最终找到error相对较小的lambda.

    相关文章

      网友评论

          本文标题:Underfitting and overfitting

          本文链接:https://www.haomeiwen.com/subject/pdevzttx.html