将 function 应用于 R 中的列表列表的最佳方法是什么？ 特别是如果内部变量都被称为相同的东西

Question

Hi I am fairly new to R and would appreciate any help on this one.

I have searched for similar questions but I unfortunately I dont really understand the solutions given.

My problem:

I have about 60 excel sheets full of repeated testing data that I want to analyse and compare. These all have a similar structure and variable/column names but the number of data points is different for each one. I have loaded these into R as a list of lists and I want to perform a series of manipulations on each original data set once it is inside the list of lists. These manipulations would be identical using the same variable names etc but applied to different data sets.

As an example say I wanted to calculate something based on the data and then add the results as a new variable inside the nested list.

A simplified version of my situation would be something like this.

###set up###

specimen1=list("Stress"=50:100,
"Strain"=5:55) #represents my excel sheet imports
specimen2=list("Stress"=65:115,
"Strain"=6.5:56.5) #simplifed for brevity

specimen3=list("Stress"=42:92,"Strain"=4.2:54.2)

rate1=list(specimen1,specimen2,specimen3) #my list of lists

names(rate1)<-c("specimen 1","specimen 2","specimen 3") #set the names

####performing calculation and adding to the list entry###

#now i want to perform a calculation on each specimen and then add the result to that specimen
#I suspect the solution lies with the lapply family something like this?

example_function<-function(Stress,Strain){

E=Stress/Strain #performs calculation
#but doesn't add the result to the list?

rate1$specimen$E=E #something like this to add to the original data set?
#but I don't understand how to change the indexing with out using a for loop

}


lapply(rate1,example_function)

#########

What is the best way to perform a function on each element of list of lists which then adds a variable to all those list components?

I suspect that the solution to this will be simple?

Answer 1

If you are not tied to doing this in lists , you can bind all your lists and do it in the resulting data.frame format using dplyr

library(dplyr)
bind_rows(rate1, .id="specimen") %>% 
 mutate(E = Stress/Strain)

which produces

# A tibble: 153 x 4
   specimen   Stress Strain     E
   <chr>       <int>  <dbl> <dbl>
 1 specimen 1     50      5 10   
 2 specimen 1     51      6  8.5 
 3 specimen 1     52      7  7.43
 4 specimen 1     53      8  6.62
...

Using data.frames is usually the most straightforward way of doing things in R .

If you want to keep using lists , because you have to make a new column, it's possibly better to use a for loop instead of lapply . This is how you solve your particular issue using a loop.

# this will add the column E to each element of the list rate1
for(i in 1:length(rate1)) {
  rate1[[i]]$E <- rate1[[i]]$Stress/rate1[[i]]$Strain  
}

This is the version with lapply , you can keep adding columns inside the list call of the function(li) .

modified_rate1 <- 
lapply(rate1, function(li) 
  list(
    Stress = li$Stress,
    Strain = li$Strain,
    E = li$Stress/li$Strain
  )
)

I think doing this in a data.frame is the way to go, but you should see what's better for your other many purposes