How to get rid of warnings about NAs?

[erinho]

I'm practicing plot making, and I have some issue with NAs. I get two warning messages:

1. Expected 2 pieces. Missing pieces filled with NA in 605 rows [189, 268, etc...
2. In between(hour, 10, 12) : NAs introduced by coercion

This occurs even though I ran the code below before plotting. So how can I solve this NA problem?

df <- df %>% 
  na.omit()

The plot ↓

df %>% 
  separate(Time, c("hour", "minutes"), sep = ":", convert = TRUE) %>% 
  filter(between(hour, 10, 12)) %>% 
  ggplot() +
  aes(hour) +
  stat_count()

Sample data

structure(list(Date = c("9.3.2019", "5.9.2019", "23.12.2019", 
"8.6.2018", "16.1.2019", "19.12.2018", "25.1.2019", "9.10.2014", 
"11.3.2020", "8.5.2017", "9.12.2019", "17.10.2016", "7.12.2018", 
"24.6.2017", "1.10.2018", "24.2.2016", "15.3.2019", "6.3.2020", 
"3.12.2019", "20.1.2020", "25.2.2015", "20.10.2018", "28.3.2019", 
"13.6.2019", "12.5.2016", "8.9.2017", "23.11.2016", "24.7.2015", 
"25.2.2014", "21.7.2019", "22.7.2013", "4.12.2014", "17.3.2016", 
"25.9.2017", "2.12.2016", "23.7.2012", "15.9.2016", "1.1.2018", 
"29.3.2018", "9.11.2017", "7.12.2017", "13.12.2012", "27.12.2016", 
"1.6.2019", "9.12.2012", "14.11.2018", "14.3.2013", "25.8.2016", 
"30.9.2015", "23.6.2015", "3.7.2019", "28.12.2013", "13.3.2014", 
"27.5.2019", "16.5.2016", "10.11.2019", "2.12.2014", "14.12.2015", 
"15.2.2018", "1.2.2015", "16.5.2015", "26.10.2018", "3.1.2016", 
"6.3.2014", "16.11.2015", "9.3.2015", "17.7.2019", "19.5.2017", 
"17.9.2014", "19.2.2017", "26.8.2013", "19.3.2013", "29.3.2019", 
"14.3.2019", "25.1.2015", "11.11.2013", "16.6.2013", "11.2.2018", 
"5.3.2019", "2.7.2017", "10.4.2015", "20.1.2015", "27.7.2017", 
"4.9.2013", "7.9.2016", "16.12.2016", "26.3.2018", "17.12.2014", 
"17.8.2012", "20.4.2013", "4.12.2014", "28.5.2014", "1.6.2012", 
"20.1.2020", "11.9.2015", "16.8.2019", "14.11.2014", "24.1.2015", 
"13.11.2015", "8.2.2020"), Time = c("19:00", "10:00", "17:40", 
"15:00", "16:20", "09:10", "13:00", "-", "19:30", "13:00", "-", 
"09:50", "07:30", "20:00", "10:00", "16:40", "12:40", "12:10", 
"08:30", "14:30", "11:40", "15:00", "09:20", "10:30", "09:30", 
"13:50", "16:40", "18:10", "13:40", "10:00", "-", "13:00", "14:40", 
"08:00", "10:00", "16:20", "09:20", "15:30", "08:20", "12:00", 
"13:30", "16:10", "07:50", "16:20", "19:50", "13:10", "07:10", 
"19:30", "08:30", "22:50", "17:40", "18:00", "08:30", "10:20", 
"13:00", "20:10", "12:10", "10:00", "14:00", "11:20", "08:00", 
"11:40", "11:40", "09:30", "17:30", "00:00", "18:00", "08:10", 
"15:00", "18:30", "13:10", "18:00", "09:00", "08:10", "08:00", 
"16:40", "18:00", "19:10", "13:00", "14:00", "09:10", "10:20", 
"19:40", "12:30", "09:10", "11:30", "13:50", "10:10", "12:10", 
"11:30", "19:30", "12:30", "19:40", "09:20", "13:20", "09:30", 
"11:20", "19:20", "08:40", "07:30")), row.names = c(NA, -100L
), class = c("tbl_df", "tbl", "data.frame"))

[andresrcs]

@erinho

Missing pieces filled with NA in 605 rows [189, 268, etc...

This occurs because in some rows you have a "-" instead of a time value, so it can't be separated into hours and minutes, if you want to avoid this simply filter out the rows without time information.

[erinho]

And how to do that? This code didn't do the job

df= df[complete.cases(df),]

[Riffomonas]

Hey @erinho -

A couple of things to try. First, you can use filter to remove those values in Time that are only a hyphen...

df %>% filter(Time != "-")

Alternatively, if you want to drop the rows with NA values after you use separate, you can use drop_na...

df %>% 
  separate(Time, c("hour", "minutes"), sep = ":", convert = TRUE) %>%
  drop_na()

You'll still get a warning from the separate command. If that bugs you, then you should probably use filter first

df %>%
  filter(Time != "-") %>%
  separate(Time, c("hour", "minutes"), sep = ":", convert = TRUE)

Your efforts with na.omit didn't work because to that point, df hadn't been parsed to split up the dates.

Another thought, if you read these values in from a file, you can use read_tsv or read_csv with na = c("", "NA", "-") to automatically turn the - to an NA value from the outset.

[gitdemont]

Hi,

Using suppressWarnings() ?

[erinho]

Lovely, thanks Riffomonas