• Python scikit-learn (metrics): difference between r2_score and explained_variance_score?


    I noticed that that 'r2_score' and 'explained_variance_score' are both build-in sklearn.metrics methods for regression problems.

    I was always under the impression that r2_score is the percent variance explained by the model. How is it different from 'explained_variance_score'?

    When would you choose one over the other?

    Thanks!

    OK, look at this example:

    In [123]:
    #data
    y_true = [3, -0.5, 2, 7]
    y_pred = [2.5, 0.0, 2, 8]
    print metrics.explained_variance_score(y_true, y_pred)
    print metrics.r2_score(y_true, y_pred)
    0.957173447537
    0.948608137045
    In [124]:
    #what explained_variance_score really is
    1-np.cov(np.array(y_true)-np.array(y_pred))/np.cov(y_true)
    Out[124]:
    0.95717344753747324
    In [125]:
    #what r^2 really is
    1-((np.array(y_true)-np.array(y_pred))**2).sum()/(4*np.array(y_true).std()**2)
    Out[125]:
    0.94860813704496794
    In [126]:
    #Notice that the mean residue is not 0
    (np.array(y_true)-np.array(y_pred)).mean()
    Out[126]:
    -0.25
    In [127]:
    #if the predicted values are different, such that the mean residue IS 0:
    y_pred=[2.5, 0.0, 2, 7]
    (np.array(y_true)-np.array(y_pred)).mean()
    Out[127]:
    0.0
    In [128]:
    #They become the same stuff
    print metrics.explained_variance_score(y_true, y_pred)
    print metrics.r2_score(y_true, y_pred)
    0.982869379015
    0.982869379015

    So, when the mean residue is 0, they are the same. Which one to choose dependents on your needs, that is, is the mean residue suppose to be 0?

    Most of the answers I found (including here) emphasize on the difference between R2 and Explained Variance Score, that is: The Mean Residue (i.e. The Mean of Error).

    However, there is an important question left behind, that is: Why on earth I need to consider The Mean of Error?


    Refresher:

    R2: is the Coefficient of Determination which measures the amount of variation explained by the (least-squares) Linear Regression.

    You can look at it from a different angle for the purpose of evaluating the predicted values of y like this:

    Varianceactual_y × R2actual_y = Variancepredicted_y

    So intuitively, the more R2 is closer to 1, the more actual_y and predicted_y will have samevariance (i.e. same spread)


    As previously mentioned, the main difference is the Mean of Error; and if we look at the formulas, we find that's true:

    R2 = 1 - [(Sum of Squared Residuals / n) / Variancey_actual]
    
    Explained Variance Score = 1 - [Variance(Ypredicted - Yactual) / Variancey_actual]

    in which:

    Variance(Ypredicted - Yactual) = (Sum of Squared Residuals - Mean Error) / n 

    So, obviously the only difference is that we are subtracting the Mean Error from the first formula! ... But Why?


    When we compare the R2 Score with the Explained Variance Score, we are basically checking the Mean Error; so if R2 = Explained Variance Score, that means: The Mean Error = Zero!

    The Mean Error reflects the tendency of our estimator, that is: the Biased v.s Unbiased Estimation.


    In Summary:

    If you want to have unbiased estimator so our model is not underestimating or overestimating, you may consider taking Mean of Error into account.

    参考链接:https://stackoverflow.com/questions/24378176/python-sci-kit-learn-metrics-difference-between-r2-score-and-explained-varian

  • 相关阅读:
    异常处理
    面向对象 -- 内置方法
    面向对象 -- 反射(详细)
    面向对象 -- 反射
    面向对象 -- 三大特性之封装
    pickle模块 collections模块在面向对象中的应用
    面向对象 -- 三大特性之多态
    Third Week(补充完整)
    Second Week(补充完整)
    First Week (补充完整)
  • 原文地址:https://www.cnblogs.com/jingsupo/p/9847907.html
Copyright © 2020-2023  润新知