过拟合和欠拟合

截取转载自kaggle 教程，https://www.kaggle.com/dansbecker/underfitting-and-overfitting

结论

• Overfitting: capturing spurious patterns that won't recur in the future, leading to less accurate predictions,
  过拟合：获取了在不会再次发生的虚假的模式，导致预测率降低
• Underfitting: failing to capture relevant patterns, again leading to less accurate predictions.
  欠拟合：没有获取到相关的模式，导致预测率降低

image.png