minimum value in dataframe greater than 0 in R-CodePudding

I have a dataset with ~2500 columns in R, and I am trying to find the minimum value greater than zero from the entire data frame. Once I have found this number, I want to try adding it to all other data in the data frame.

Here is an example dataset.

df <- as.data.frame(rbind(c(0.01,0.4,0.5,5,27,0.22,25,0,0,0.3),c(0.25,0,0,2.3,3.6,6,0,0.001,0.021,0),c(22,23,0,40,0.53,0.2,0.4,0.44,0,0),c(0.1,0,0.12,0.56,0.7,13,0,0,3,4)))

As you can see, the minimum value in this dataframe is 0.001

I have tried the following:

min<- min(df=!0)

Which gives me 1.

I then want to add this number (0.001) to all other values in the data frame. However, I have no idea how to do this.

Thank you for your time considering my question. Any help would be greatly appreciated.

CodePudding user response：

Your syntax of df=!0 is incorrect. If you want to specify df not equal to zero, the correct way is df!=0. However, if you do it this way, the output would be logical, so the result of min(df != 0) will be 0.

Therefore, you should use the logical output to index your original df.

min(df[df != 0])
[1] 0.001

Since your question stated that you would like to have min values greater than zero, using df > 0 is more accurate.

min(df[df > 0])
[1] 0.001

CodePudding user response：

Another way is to unlist the data frame, and then find the minimum of its nonzero elements:

nonzero <- which(df != 0)
min(unlist(df)[nonzero])
[1] 0.001

CodePudding user response：

Here is another option using tidyverse with min. We can convert any of the zero values to NA using map, then convert the data to a vector, then pull the minimum value, excluding NA.

library(tidyverse)

map(df, na_if, 0) %>%
  flatten_dbl() %>%
  min(na.rm = T)

# [1] 0.001