参考：http://scikit-learn.org/stable/modules/learning_curve.html
estimator‘s generalization error can be decomposed in terms of
bias, variance and noise. The bias of
an estimator is its average error for different training sets. The variance of
an estimator indicates how sensitive it is to varying training sets. Noise is a property of the data.
具体内容有时间翻译。。。
版权声明：本文为博主原创文章，未经博主允许不得转载。
scikit-learn：3.5. Validation curves: plotting scores to evaluate models
原文：http://blog.csdn.net/mmc2015/article/details/47144197