在一个数据帧中将行（带有行名称）乘以其他列中的匹配列名称

本文介绍了在一个数据帧中将行（带有行名称）乘以其他列中的匹配列名称的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！ 问题描述

我有两个数据框：

df1 <- data.frame(Values=c(0.01,0.05), row.names=c("X", "Y")) df1 Values X 0.01 Y 0.05 df2 <-data.frame(c(0,1,1), c(1,0,0), c(1,1,1)) colnames(df2) <- c("X","Y","Z") df2 X Y Z 1 0 1 1 2 1 0 1 3 1 0 1

我希望在df2上执行横向操作，在每个列乘以列在df2中，其df1中的对应行，然后执行求和。

I wish to perform a rowwise operation on df2, where I multiply every column in df2 with its corresponding row in df1 and then perform a summation.

例如，对于df2的第1行，我想计算：

For example, for row 1 of df2, I wish to calculate:

df2 %>% rowwise %>% mutate(newVAL=(df1["X",]*df2[1,"X"])+(df1["Y",]*df2[1,"Y"]))

，同时排除不匹配的列（df1中的行）或具有NAs 。

while excluding columns that don't match (rows in df1) or have NAs.

df1中有几千行，df2中有数千行。

I have several thousands of rows in df1 and several thousands of rows and columns in df2.

任何帮助非常感谢！

PS。我已经在Perl中使用哈希实现了这一点，并且正在使用system（）调用来在Rmarkdown文档中执行这些计算。为了保持完全可重现性，我正在尝试在R中重做。如果需要，快乐分享Perl代码。

PS. I have implemented this in Perl using hashes and was using the system() call to perform these calculations within an Rmarkdown document. In order to keep it completely reproducible, I am trying to redo it in R. Happy to share the Perl code if necessary.

谢谢。

推荐答案

如果我理解正确，看起来您需要 sweep 。

If I understand correctly, it looks like you need sweep.

df3 <- sweep(df2[, rownames(df1)], 2, t(df1), '*') df3$total <- rowSums(df3)