当data.frame读取为非数字时，如何在条形图上记录比例y轴？ [R

Question

I am having issues with the argument log = "y" from barplot()

        ADR BCLREQ CALL  BOND FUT     CD   CDS CDX 
IO Only    "0"   "3"         "0"      "0"  "0" "1" 
IO and TS  "0"   "0"         "0"      "0"  "0" "0" 
No Changes "5"   "45"        "9"      "39" "2" "11"
TS Only    "0"   "0"         "0"      "0"  "0" "0" 
Freq       "5"   "48"        "9"      "39" "2" "12".  

.

 > dput(data)
structure(c("0", "0", "5", "0", "5", "3", "0", "45", "0", "48", 
"0", "0", "9", "0", "9", "0", "0", "39", "0", "39", "0", "0", 
"2", "0", "2", "1", "0", "11", "0", "12"), .Dim = 5:6, .Dimnames = list(
    c("IO Only", "IO and TS", "No Changes", "TS Only", "Freq"
    ), c("ADR", "BCLREQ CALL", "BOND FUT", "CD", "CDS", "CDX"
    )))

Now the barplot

  BAR2 <- barplot(data[1:4,], main = "Build Efficiency", ylab = "Manual Frequency", 
                              xlab = "IO Screen", beside = F, log = "y",
                              col = c("blue", "green", "red", "yellow"), las = 1, legend.text = rownames(data)[1:4])

The above should result in:

I would like to make the 'Y' axis in log scale. after adding log = "y", I get:

Error in height + offset : non-numeric argument to binary operator

After a bit of digging I realized my data might not be numeric? not sure if checking correctly but:

> str(data[1,2])
 chr "3"

So I tried:

  data <- data.matrix(data, rownames.force = NA)

  data[1:5,] <- sapply(data[1:5,], as.numeric)

I still get:

Error in height + offset : non-numeric argument to binary operator

Answer 1

As the comments mentioned, your data is stored as characters. Since data is a matrix, use mode(data) = "numeric" .

You also have the problem that 0's are in your dataset and you cannot log transform this without a transformation. See this post for some solutions: https://stats.stackexchange.com/questions/1444/how-should-i-transform-non-negative-data-including-zeros