Simple tricks for `rstats`

In my Twitter account, I usually post some simple R tricks, things that come up when I’m working and don’t cost much to share in 2 minutes, in case I can help someone. I just realized that they are really useful and that having them scattered on Twitter is a problem, so I thought I’d collect them in a blog post to make them easier to find (even for me). Here are some of those tricks:

Pass data from a data.frame to the clipboard, useful when you want to move data from R to Excel without complications:

write.table(borra, "clipboard", sep = "\t", dec = ",", row.names = F)

Pass the field names of a data.frame to the clipboard (connects with the previous one), useful when you work with a text editor or a spreadsheet to automate code:

write.table(colnames(DF), "clipboard", sep = "\t", dec = ",", row.names = F)

Set format 00000 for postal codes:

cp <- c(8080, 29001)
cp <- sprintf("%05d", cp)

The best subset for H2O:

df.hex[df.hex$campo > 0, ]

Text to date in R:

dffecha = as.Date(dffecha, "%d/%m/%Y")

Identify repeated records in a data.frame; creates a data.frame with duplicate records in one line of dplyr code:

repetidos <- df %>%
  group_by(campo_ID) %>%
  summarise(repetido = n()) %>%
  filter(repetido > 1)

My favorite and the reason for the post, binning a quantitative variable into $n$ groups:

grupos = 10
df <- df %>%
  arrange(campo) %>%
  mutate(campo_tramos = as.factor(ceiling((row_number() / n()) * grupos)))

Transform nulls to 0 in 20 characters:

df[is.na(df$V1)] <- 0

Transform numbers separated by a comma in text format to numeric format:

dfnumero <- as.numeric(sub(",", ".", dftexto))

All factors of my R data.frame to character to avoid some trouble, using lapply:

df <- data.frame(lapply(df, as.character), stringsAsFactors = FALSE)

Create a sequence of dates in R and format the sequence of dates in R:

secuencia <- seq(as.Date("2016/1/1"), by = "month", length.out = 12)
format(secuencia, "%Y%m")

Function to get only the numbers that start a text string in R:

library(readr)
parse_number("1234abcd")

Using tidyr to extract only numbers from a string. Note that sometimes you need talent to do this task:

library(tidyr)
extract_numeric("abcd1234efgh")

Function to export larger dataframes to the clipboard, ideal for moving dataframes to Excel, converting from American to European format:

write_excel <- function(x, row.names = FALSE, col.names = TRUE, ...) {
  write.table(x, "clipboard-16384", sep = "\t", dec = ",", row.names = row.names, col.names = col.names, ...)
}

Sequence of dates with R:

seq(as.Date("2018-01-01"), as.Date("2018-12-31"), by = "days")

Function to get only numbers within a text string in R; there are many, I use this one:

gsub("[^0-9]", "", "abc123def")

The not in in R:

`%nin%` <- Negate(`%in%`)

Allow RStudio to display all columns of the data.frame:

options(max.print = 10000)

Replace a variable with a null value to 0 using ifelse:

df$var <- ifelse(is.na(df$var), 0, df$var)

Operate with months in YYYYMM format, the typical ones for partitions, a trick that probably already exists:

mes <- 201812
mes_sig <- as.numeric(format(as.Date(paste0(mes, "01"), "%Y%m%d") + 31, "%Y%m"))

I hope this post can continue to grow, they are trifles (my trifles) and I have them centralized in a single post.