Question

我创建了一个函数，该函数返回分数并打印有关结果的消息。这里的代码：

def compute_score(X_cv, clf):
    score = clf.predict_proba(X_cv[84582].reshape(1,-1))[0][1] # Prob of a Win
    df_top10_feat = pd.DataFrame(data={"Feature":df_cv.columns[:-1],
                             "Coefficient":clf.coef_[0],
                             "Value":X_cv[84582],
                             "Importance":list(clf.coef_ * X_cv[84582])[0]}). \
                                    sort_values("Importance",
                                                ascending=False)

    ##extract positif importance
    df_top10_feat_positif= df_top10_feat[df_top10_feat['Importance']>0]

    #extract negatif importance
    df_top10_feat_negatif= df_top10_feat[df_top10_feat['Importance']<0]



    #print
    print("The average of ", df_top10_pos_sort['Feature'].iloc[0], "is",  format(df_cv[df_top10_pos_sort['Feature'].iloc[1]].mean(), '.3f'),". The", df_top10_pos_sort['Feature'].iloc[0], "for this opportunity line is", format(df_top10_pos_sort['Value'].iloc[0], '.3f'), "Therefore, the", df_top10_pos_sort['Feature'].iloc[0],  "is lower than other similar opportunity lines.")
    print("The average of", df_top10_pos_sort['Feature'].iloc[1], "is", format(df_cv[df_top10_pos_sort['Feature'].iloc[1]].mean(), '.3f'),". The ",df_top10_pos_sort['Feature'].iloc[1], "for this opportunity line is", format(df_top10_pos_sort['Value'].iloc[1], '.3f'), "Therefore, the", df_top10_pos_sort['Feature'].iloc[1], " is lower than other similar opportunity lines.")
    print("The average of", df_top10_pos_sort['Feature'].iloc[2], "is", format(df_cv[df_top10_pos_sort['Feature'].iloc[2]].mean(), '.3f'),". The", df_top10_pos_sort['Feature'].iloc[1], "for this opportunity line is", format(df_top10_pos_sort['Value'].iloc[2], '.3f'), "Therefore, the", df_top10_pos_sort['Feature'].iloc[2], "is lower than other similar opportunity lines.")
    print("The average of", df_top10_neg_sort['Feature'].iloc[0], "is", format(df_cv[df_top10_neg_sort['Feature'].iloc[0]].mean(), '.3f'), ". The",df_top10_neg_sort['Feature'].iloc[0], "for this opportunity line is", format(df_top10_neg_sort['Value'].iloc[0], '.3f'), "Therefore,", df_top10_neg_sort['Feature'].iloc[0], "is lower than other similar opportunity lines.")
    print("The average of", df_top10_neg_sort['Feature'].iloc[1], "is", format(df_cv[df_top10_neg_sort['Feature'].iloc[1]].mean(), '.3f'),". The", df_top10_neg_sort['Feature'].iloc[1],  "for this opportunity line is", format(df_top10_neg_sort['Value'].iloc[1], '.3f'), "Therefore,", df_top10_neg_sort['Feature'].iloc[1], "is lower than other similar opportunity lines.")
    print("The average of", df_top10_neg_sort['Feature'].iloc[2],"is", format(df_cv[df_top10_neg_sort['Feature'].iloc[2]].mean(), '.3f'),". The",  df_top10_neg_sort['Feature'].iloc[2], "for this opportunity line is", format(df_top10_neg_sort['Value'].iloc[2], '.3f'), "Therefore,", df_top10_neg_sort['Feature'].iloc[2], "is lower than other similar opportunity lines.")
    return

我的问题是我应该在“返回”集团中添加一个假设吗？还是我这样保留它？

谢谢

Answer 1

如果要获得分配给变量的结果，则应在返回行中添加一些内容；否则，请清除return关键字。

Answer 2

这取决于。如果您想使用（计算出的）结果，那么可以，您应该以某种格式返回它（例如，结果的dict）如果打印足以满足您的需要，则可以按原样保留它。

[编辑：在This answer之后，我认为您确实应该擦除return。从功能上讲，它没有什么区别，但可以帮助读者理解，该函数的确在任何情况下都不打算返回任何内容（仅打印）。

用python计算分数的函数

2 个答案: