Working with R often involves encountering various issues. Here are some collected tips.

Data Frame Operations

Extracting a Single Column/Row Without Returning a Vector

• When using df[] to subset a data frame, if you extract a single column or row, or delete all but one column or row, the resulting object will be a vector instead of a data frame. You need to specify the parameter to ensure it returns a data frame.
• This method also applies to matrices.

1

df[c(1, 2, 3), -2, drop = F]

Merging Data Using row.names as Key

• Many Chinese materials describe merging data by using one column as the key for row-wise merging. However, you can also use row.names as the key, but after merging, row.names will become a column of data. You need to specify row.names again.

1
2
3
4
5

# Two-step process
merge_data <- merge(x, y, all.x = T, by = 0)
merge_data <- data.frame(merge_data), row.names = 1)
# One-step merge
merge_data <- data.frame(merge(x, y, all.x = T, by = 0), row.names = 1)

Other

R’s Python Dict Alternative

• Many online resources use lists combined with the match function to implement dictionary-like functionality. However, in R, named vectors can be used to achieve key-value mapping, which is very convenient. However, it’s unclear whether this approach is efficient.

1
2
3
4

a <- c(1, 2, 3)
names(a) <- c("a", "b", "c")
tmp <- a["a"]
print(tmp)