recode

In [1]:

# https://dplyr.tidyverse.org/reference/recode.html
%run nb_helpers.py

from datar.all import *

nb_header(recode)

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

Try this notebook on binder.

★ recode

¶

Recode a vector, replacing elements in it¶

Args:¶

  x: A vector to modify
  *args: and
  **kwargs: replacements
  _default: If supplied, all values not otherwise matched will be
    given this value. If not supplied and if the replacements are
    the same type as the original values in series, unmatched values
    are not changed. If not supplied and if the replacements are
    not compatible, unmatched values are replaced with np.nan.

  _missing: If supplied, any missing values in .x will be replaced
    by this value.

Returns:¶

  The vector with values replaced

In [2]:

char_vec = sample(c("a", "b", "c"), 10, replace=True)
recode(char_vec, a="Apple")

char_vec = sample(c("a", "b", "c"), 10, replace=True) recode(char_vec, a="Apple")

Out[2]:

0    Apple
1        c
2        c
3    Apple
4        b
5    Apple
6        b
7        b
8        b
9        c
dtype: object

In [3]:

recode(char_vec, a="Apple", b="Banana")

recode(char_vec, a="Apple", b="Banana")

Out[3]:

0     Apple
1         c
2         c
3     Apple
4    Banana
5     Apple
6    Banana
7    Banana
8    Banana
9         c
dtype: object

In [4]:

recode(char_vec, a="Apple", b="Banana", _default = NA)

recode(char_vec, a="Apple", b="Banana", _default = NA)

Out[4]:

0     Apple
1       NaN
2       NaN
3     Apple
4    Banana
5     Apple
6    Banana
7    Banana
8    Banana
9       NaN
dtype: object

In [5]:

level_key = dict(a="apple", b="banana", c="carrot")
recode(char_vec, **level_key)

level_key = dict(a="apple", b="banana", c="carrot") recode(char_vec, **level_key)

Out[5]:

0     apple
1    carrot
2    carrot
3     apple
4    banana
5     apple
6    banana
7    banana
8    banana
9    carrot
dtype: object

In [6]:

num_vec = c(range(4), NA)
recode(num_vec, {1: 20, 3: 40})

num_vec = c(range(4), NA) recode(num_vec, {1: 20, 3: 40})

[2022-12-02 14:19:25][datar][WARNING] Unreplaced values treated as NA as `_x` is not compatible. Please specify replacements exhaustively or supply `_default`

Out[6]:

0    NaN
1     20
2    NaN
3     40
4    NaN
dtype: object

In [7]:

recode(num_vec, "a", "b", "c", "d")

recode(num_vec, "a", "b", "c", "d")

Out[7]:

0      a
1      b
2      c
3      d
4    NaN
dtype: object

In [8]:

recode(c(0,4,2), "a", "b", "c", "d", _default="nothing")

recode(c(0,4,2), "a", "b", "c", "d", _default="nothing")

Out[8]:

0          a
1    nothing
2          c
dtype: object

In [9]:

recode(num_vec, {1: "b", 3: "d"})

recode(num_vec, {1: "b", 3: "d"})

[2022-12-02 14:19:31][datar][WARNING] Unreplaced values treated as NA as `_x` is not compatible. Please specify replacements exhaustively or supply `_default`

Out[9]:

0    NaN
1      b
2    NaN
3      d
4    NaN
dtype: object

In [10]:

recode(num_vec, "a", "b", "c", _default="other")

recode(num_vec, "a", "b", "c", _default="other")

Out[10]:

0        a
1        b
2        c
3    other
4      NaN
dtype: object

In [11]:

recode(num_vec, "a", "b", "c", _default="other", _missing="missing")

recode(num_vec, "a", "b", "c", _default="other", _missing="missing")

Out[11]:

0          a
1          b
2          c
3      other
4    missing
dtype: object

In [12]:

factor_vec = factor(c("a", "b", "c"))
factor_vec

factor_vec = factor(c("a", "b", "c")) factor_vec

Out[12]:

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

In [13]:

# categories lost
recode(factor_vec, a="Apple")

# categories lost recode(factor_vec, a="Apple")

Out[13]:

0    Apple
1        b
2        c
dtype: category
Categories (3, object): ['Apple', 'b', 'c']

In [14]:

recode_factor(factor_vec, a="Apple")

recode_factor(factor_vec, a="Apple")

Out[14]:

0    Apple
1        b
2        c
dtype: category
Categories (3, object): ['Apple', 'b', 'c']

In [15]:

recode_factor(num_vec, {0: "z", 1: "y", 2: "x"})

recode_factor(num_vec, {0: "z", 1: "y", 2: "x"})

[2022-12-02 14:19:42][datar][WARNING] Unreplaced values treated as NA as `_x` is not compatible. Please specify replacements exhaustively or supply `_default`

Out[15]:

0      z
1      y
2      x
3    NaN
4    NaN
dtype: category
Categories (3, object): ['z', 'y', 'x']

In [16]:

recode_factor(num_vec, {0: "z", 1: "y", 2: "x"}, _default="D")

recode_factor(num_vec, {0: "z", 1: "y", 2: "x"}, _default="D")

Out[16]:

0      z
1      y
2      x
3      D
4    NaN
dtype: category
Categories (4, object): ['z', 'y', 'x', 'D']

In [17]:

recode_factor(num_vec, {0: "z", 1: "y", 2: "x"}, _default="D", _missing="M")

recode_factor(num_vec, {0: "z", 1: "y", 2: "x"}, _default="D", _missing="M")

Out[17]:

0    z
1    y
2    x
3    D
4    M
dtype: category
Categories (5, object): ['z', 'y', 'x', 'D', 'M']

In [18]:

recode_factor(list(letters[:3]), b="z", c="y")

recode_factor(list(letters[:3]), b="z", c="y")

Out[18]:

0    a
1    z
2    y
dtype: category
Categories (3, object): ['z', 'y', 'a']

In [19]:

level_key = dict(a="apple", b="banana", c="carrot")
recode_factor(char_vec, **level_key)

level_key = dict(a="apple", b="banana", c="carrot") recode_factor(char_vec, **level_key)

Out[19]:

0     apple
1    carrot
2    carrot
3     apple
4    banana
5     apple
6    banana
7    banana
8    banana
9    carrot
dtype: category
Categories (3, object): ['apple', 'banana', 'carrot']