drop_na

In [1]:

# https://tidyr.tidyverse.org/reference/drop_na.html
%run nb_helpers.py

from datar.all import *

nb_header(drop_na)

# https://tidyr.tidyverse.org/reference/drop_na.html %run nb_helpers.py from datar.all import * nb_header(drop_na)

Try this notebook on binder.

★ drop_na

¶

Drop rows containing missing values¶

See https://tidyr.tidyverse.org/reference/drop_na.html

Args:¶

  data: A data frame.
  *columns: Columns to inspect for missing values.
  _how: How to select the rows to drop
    - all: All columns of columns to be NAs

    - any: Any columns of columns to be NAs

    (tidyr doesn't support this argument)

Returns:¶

  Dataframe with rows with NAs dropped and indexes dropped

In [2]:

df = tibble(x = c(1, 2, NA), y = c("a", NA, "b"))
df >> drop_na()

df = tibble(x = c(1, 2, NA), y = c("a", NA, "b")) df >> drop_na()

Out[2]:

	x	y
	<float64>	<object>
0	1.0	a

In [3]:

df >> drop_na(f.x)

df >> drop_na(f.x)

Out[3]:

	x	y
	<float64>	<object>
0	1.0	a
1	2.0	NaN

In [4]:

vars = ["y"]
df >> drop_na(f.x, any_of(vars))

vars = ["y"] df >> drop_na(f.x, any_of(vars))

Out[4]:

	x	y
	<float64>	<object>
0	1.0	a

In [5]:

# how_='any' or how_='all'
# not supported by tidyr
df >> drop_na(how_='all')

# how_='any' or how_='all' # not supported by tidyr df >> drop_na(how_='all')

Out[5]:

	x	y
	<float64>	<object>
0	1.0	a
1	2.0	NaN
2	NaN	b

In [6]:

df >> drop_na(how_='any')

df >> drop_na(how_='any')

Out[6]:

	x	y
	<float64>	<object>
0	1.0	a