forcats_lvl_addrm

In [1]:

%run nb_helpers.py
from datar.all import *

nb_header(
    fct_expand,
    fct_explicit_na,
    fct_drop,
    fct_unify,
    book="forcat_lvl_addrm",
)

%run nb_helpers.py from datar.all import * nb_header( fct_expand, fct_explicit_na, fct_drop, fct_unify, book="forcat_lvl_addrm", )

Try this notebook on binder.

★ fct_expand

¶

Add additional levels to a factor¶

Args:¶

  _f: A factor
  *additional_levels: Additional levels to add to the factor.
    Levels that already exist will be silently ignored.

Returns:¶

  The factor with levels expanded

★ fct_explicit_na

¶

Make missing values explicit¶

This gives missing values an explicit factor level, ensuring that they
appear in summaries and on plots.

Args:¶

  _f: A factor
  na_level: Level to use for missing values.
    This is what NAs will be changed to.

Returns:¶

  The factor with explict na_levels

★ fct_drop

¶

Drop unused levels¶

Args:¶

  _f: A factor
  only: A character vector restricting the set of levels to be dropped.
    If supplied, only levels that have no entries and appear in
    this vector will be removed.

Returns:¶

  The factor with unused levels dropped

★ fct_unify

¶

Unify the levels in a list of factors¶

Args:¶

  fs: A list of factors
  levels: Set of levels to apply to every factor. Default to union
    of all factor levels

Returns:¶

  A list of factors with the levels expanded

fct_expand¶

In [2]:

fct = factor(sample(letters[:3], 20, replace = TRUE))
fct

fct = factor(sample(letters[:3], 20, replace = TRUE)) fct

Out[2]:

['c', 'c', 'b', 'c', 'a', ..., 'b', 'c', 'b', 'b', 'c']
Length: 20
Categories (3, object): ['a', 'b', 'c']

In [3]:

fct_expand(fct, "d", "e", "f")

fct_expand(fct, "d", "e", "f")

Out[3]:

['c', 'c', 'b', 'c', 'a', ..., 'b', 'c', 'b', 'b', 'c']
Length: 20
Categories (6, object): ['a', 'b', 'c', 'd', 'e', 'f']

In [4]:

fct_expand(fct, letters[:6])

fct_expand(fct, letters[:6])

Out[4]:

['c', 'c', 'b', 'c', 'a', ..., 'b', 'c', 'b', 'b', 'c']
Length: 20
Categories (6, object): ['a', 'b', 'c', 'd', 'e', 'f']

fct_explicit_na¶

In [5]:

f1 = factor(c("a", "a", NA, NA, "a", "b", NA, "c", "a", "c", "b"))
fct_count(f1)

f1 = factor(c("a", "a", NA, NA, "a", "b", NA, "c", "a", "c", "b")) fct_count(f1)

Out[5]:

	f	n
	<category>	<float64>
0	a	4.0
1	b	2.0
2	c	2.0
3	NaN	3.0

In [6]:

f2 = fct_explicit_na(f1)
fct_count(f2)

f2 = fct_explicit_na(f1) fct_count(f2)

Out[6]:

	f	n
	<category>	<int64>
0	a	4
1	b	2
2	c	2
3	(Missing)	3

fct_drop¶

In [7]:

fct = factor(c("a", "b"), levels = c("a", "b", "c"))
fct

fct = factor(c("a", "b"), levels = c("a", "b", "c")) fct

Out[7]:

['a', 'b']
Categories (3, object): ['a', 'b', 'c']

In [8]:

fct_drop(fct)

fct_drop(fct)

Out[8]:

['a', 'b']
Categories (2, object): ['a', 'b']

In [9]:

fct_drop(fct, only = "a")

fct_drop(fct, only = "a")

Out[9]:

['a', 'b']
Categories (3, object): ['a', 'b', 'c']

In [10]:

fct_drop(fct, only = "c")

fct_drop(fct, only = "c")

Out[10]:

['a', 'b']
Categories (2, object): ['a', 'b']

fct_unify¶

In [11]:

fs = [factor("a"), factor("b"), factor(c("a", "b"))]
fct_unify(fs)

fs = [factor("a"), factor("b"), factor(c("a", "b"))] fct_unify(fs)

Out[11]:

[['a']
 Categories (2, object): ['a', 'b'],
 ['b']
 Categories (2, object): ['a', 'b'],
 ['a', 'b']
 Categories (2, object): ['a', 'b']]