CODEWITHSUNDEEP
r
Speldosa
Speldosa

Reputation: 1970

How can you visualize data frames in a good way?

Given that you have a data frame with a lot of columns and rows, how can you visualize this in a good way?

I have imported my data from excel, where I in a clear way can browse my data. However, when I import it as a data frame into R, things get more complicated. I quickly get lost in the output in my terminal window. Could I output my data frame into some form of more easily accessible form for viewing, something that looks more the way it would look in excel?

Upvotes: 5

Views: 3733

Answers (3)

robertspierre
robertspierre

Reputation: 4371

You can use DT, an R interface to the DataTables library.

Here is a quick function to visualize a dataframe in your browser:

viewdf <- function(df,
                   to_truncate = NULL,
                   truncate_max = 10,
                   truncate_max_title = 200,
                   types_in_colnames = TRUE,
                   filter = "top",
                   to_round = NULL,
                   to_round_digits = 2) {
  # column names to display
  newcolnames <- colnames(df)
  for (i in seq(1, length(newcolnames))) {
    newcolnames[i] <- stringr::str_replace_all(newcolnames[i], "_", " ")
  }
  if (types_in_colnames) {
    for (i in seq(1, length(colnames(df)))) {
      col <- colnames(df)[i]
      newcolnames[i] <- paste0(
        newcolnames[i],
        "\n{",
        class(df[[col]]),
        "}"
      )
    }
  }
  # We need to replace NA values with a character value
  # for columns we are going to truncate,
  # otherwise DT crashes and show an empty table
  to_truncate_ix <- which(colnames(df) %in% to_truncate)
  l <- list()
  for (el in to_truncate) {
    l[[el]] <- "NA"
  }
  df %<>% tidyr::replace_na(l)
  # Generate DT table
  dtobj <- DT::datatable(df,
    colnames = newcolnames,
    filter = filter,
    options = list(
      pageLength = 20,
      filter = "top",
      autoWidth = TRUE,
      columnDefs = list(list(
        targets = to_truncate_ix,
        render = DT::JS(
          "function(data, type, row, meta) {",
          "return type === 'display' && ",
          "data.length > ", truncate_max, " ?",
          "'<span title=\"' + ",
          "data.substr(0,", truncate_max_title, ") + ",
          "'...\">' + ",
          "data.substr(0, ", truncate_max, ") + ",
          "'...</span>' : data;",
          "}"
        )
      ))
    )
  )
  # Round number columns
  for (col in to_round) {
    dtobj %<>% DT::formatRound(col, to_round_digits)
  }
  return(dtobj)
}

Upvotes: 0

chl
chl

Reputation: 29367

RStudio does a pretty good job with its built-in (read-only) data viewer. Other solutions have been suggested on Cross Validated: Is there a good browser/viewer to see an R dataset (.rda file).

enter image description here

Upvotes: 5

flodel
flodel

Reputation: 89057

You can use edit(my.data.frame).

It will open your data.frame into the default editor specified by getOption("editor"). You can use option(editor = ".......") to change that default, or just use the editor= option when calling edit. This is only relevant for Unix users.

Finally, I'll bring your attention to this important portion of the ?edit documentation:

It is important to realize that edit does not change the object called name. Instead, a copy of name is made and it is that copy which is changed. Should you want the changes to apply to the object name you must assign the result of edit to name. (Try fix if you want to make permanent changes to an object.)

Upvotes: 5

Related Questions