Question

我有一个命名向量：

dput(model_object$regression_feature_importance_matrix)
c(spend_7d = 0.704766041126329, d7_utility_sum = 0.168417670114147, 
recent_utility_ratio = 0.101030461314979, IOS = 0.0096061893294683, 
is_publisher_organic = 0.008247148545793, is_publisher_facebook = 0.0079324895692841

看起来像这样：

model_object$regression_feature_importance_matrix
             spend_7d        d7_utility_sum  recent_utility_ratio                   IOS  is_publisher_organic is_publisher_facebook 
          0.704766041           0.168417670           0.101030461           0.009606189           0.008247149           0.007932490

我想将其转置并制成数据框。我可以这样：

    model_object$regression_feature_importance_matrix %>% bind_rows() %>% t() %>% data.frame() %>% rownames_to_column() %>% rename(Feature = rowname)
                Feature           .
1              spend_7d 0.704766041
2        d7_utility_sum 0.168417670
3  recent_utility_ratio 0.101030461
4                   IOS 0.009606189
5  is_publisher_organic 0.008247149
6 is_publisher_facebook 0.007932490

1）是否有更优雅，更短的方法？ model_object$regression_feature_importance_matrix %>% bind_rows() %>% t() %>% data.frame() %>% rownames_to_column()

我只是想像大概是将命名向量转换为df会很常见，我想知道是否有一种现成的方式？

2）我试图重命名：

工作正常：model_object$regression_feature_importance_matrix %>% bind_rows() %>% t() %>% data.frame() %>% rownames_to_column() %>% rename(Feature = rowname)

不起作用：

model_object$regression_feature_importance_matrix %>% bind_rows() %>% t() %>% data.frame() %>% rownames_to_column() %>% rename(Feature = rowname, Gain = .)
Error in rename(Feature = rowname, Gain = .) : object 'rowname' not found

我想我知道为什么。点表示在管道运算符中传递的上一个对象，而我想引用列名以便更改它。如何更改列名“。”在同一个dplyr管道中“获得”收益？与仅names(df) <- c('Features', 'Gain')

相反

Answer 1

我们还可以使用stack将命名向量转换为数据帧，然后使用rename_all

stack(x)[2:1] %>% dplyr::rename_all(~c("Feature","Gain"))

#                Feature        Gain
#1              spend_7d 0.704766041
#2        d7_utility_sum 0.168417670
#3  recent_utility_ratio 0.101030461
#4                   IOS 0.009606189
#5  is_publisher_organic 0.008247149
#6 is_publisher_facebook 0.007932490

数据

x <- c(spend_7d = 0.704766041126329, d7_utility_sum = 0.168417670114147, 
recent_utility_ratio = 0.101030461314979, IOS = 0.0096061893294683, 
is_publisher_organic = 0.008247148545793, is_publisher_facebook = 0.0079324895692841)

使用bind_rows将命名向量转换为df

1 个答案: