context

In [1]:

# https://dplyr.tidyverse.org/reference/context.html
%run nb_helpers.py
from datar.all import *

nb_header(cur_group_id, cur_group_rows, cur_data, cur_data_all, cur_column, book='context')

# https://dplyr.tidyverse.org/reference/context.html %run nb_helpers.py from datar.all import * nb_header(cur_group_id, cur_group_rows, cur_data, cur_data_all, cur_column, book='context')

Try this notebook on binder.

★ cur_group_id

¶

Get the current group id¶

Args:¶

  _data: The dataframe

Returns:¶

  The current group id

★ cur_group_rows

¶

Get the current group row indices¶

Args:¶

  _data: The dataframe

Returns:¶

  The current group rows

★ cur_data

¶

Get the current dataframe¶

Args:¶

  _data: The dataframe

Returns:¶

  The current dataframe

★ cur_data_all

¶

Get the current data for the current group including¶

the grouping variables

Args:¶

  _data: The dataframe

Returns:¶

  The current dataframe

★ cur_column

¶

Get the current column¶

Args:¶

  _data: The dataframe
  _name: The column name

Returns:¶

  The current column

In [2]:

df = tibble(
  g=['a'] + ['b'] * 2 + ['c'] * 3,
  x=runif(6),
  y=runif(6)
)
gf = df >> group_by(f.g)

gf >> summarise(n = n())

df = tibble( g=['a'] + ['b'] * 2 + ['c'] * 3, x=runif(6), y=runif(6) ) gf = df >> group_by(f.g) gf >> summarise(n = n())

Out[2]:

	g	n
	<object>	<int64>
0	a	1
1	b	2
2	c	3

In [3]:

gf

gf

Out[3]:

	g	x	y
	<object>	<float64>	<float64>
0	a	0.879733	0.242456
1	b	0.789550	0.165711
2	b	0.073877	0.198040
3	c	0.677826	0.186310
4	c	0.324069	0.212226
5	c	0.589881	0.990174

TibbleGrouped: g (n=3)

In [4]:

gf >> mutate(id=cur_group_id())

gf >> mutate(id=cur_group_id())

Out[4]:

	g	x	y	id
	<object>	<float64>	<float64>	<int64>
0	a	0.879733	0.242456	0
1	b	0.789550	0.165711	1
2	b	0.073877	0.198040	1
3	c	0.677826	0.186310	2
4	c	0.324069	0.212226	2
5	c	0.589881	0.990174	2

TibbleGrouped: g (n=3)

In [5]:

gf >> summarise(row=cur_group_rows())

gf >> summarise(row=cur_group_rows())

Out[5]:

	g	row
	<object>	<object>
0	a	[0]
1	b	[1, 2]
2	c	[3, 4, 5]

In [6]:

gf_group = gf >> summarise(data=cur_group())
gf_group

gf_group = gf >> summarise(data=cur_group()) gf_group

Out[6]:

	g	data
	<object>	<object>
0	a	<DF 1x1>
1	b	<DF 1x1>
2	c	<DF 1x1>

In [7]:

gf_group >> pull(f.data)

gf_group >> pull(f.data)

Out[7]:

0    <DF 1x1>
1    <DF 1x1>
2    <DF 1x1>
Name: data, dtype: object

In [8]:

gf_data = gf >> summarise(data=cur_data())
gf_data

gf_data = gf >> summarise(data=cur_data()) gf_data

Out[8]:

	g	data
	<object>	<object>
0	a	<DF 1x2>
1	b	<DF 2x2>
2	c	<DF 3x2>

In [9]:

gf_data >> pull(f.data, to='list')

gf_data >> pull(f.data, to='list')

Out[9]:

[          x         y
   <float64> <float64>
 0  0.879733  0.242456,
           x         y
   <float64> <float64>
 1  0.789550  0.165711
 2  0.073877  0.198040,
           x         y
   <float64> <float64>
 3  0.677826  0.186310
 4  0.324069  0.212226
 5  0.589881  0.990174]

In [10]:

gf_data_all = gf >> summarise(data=cur_data_all())
gf_data_all

gf_data_all = gf >> summarise(data=cur_data_all()) gf_data_all

Out[10]:

	g	data
	<object>	<object>
0	a	<DF 1x3>
1	b	<DF 2x3>
2	c	<DF 3x3>

In [11]:

gf_data_all >> pull(f.data, to='list')

gf_data_all >> pull(f.data, to='list')

Out[11]:

[         g         x         y
   <object> <float64> <float64>
 0        a  0.879733  0.242456,
          g         x         y
   <object> <float64> <float64>
 1        b  0.789550  0.165711
 2        b  0.073877  0.198040,
          g         x         y
   <object> <float64> <float64>
 3        c  0.677826  0.186310
 4        c  0.324069  0.212226
 5        c  0.589881  0.990174]

In [12]:

df >> select(f.x, f.y) >> mutate(
    across(
        everything(), 
        lambda x, cc: [cc + ' '] * x.shape[0] + (x**2).astype(str), cc=cur_column()
    )
)

df >> select(f.x, f.y) >> mutate( across( everything(), lambda x, cc: [cc + ' '] * x.shape[0] + (x**2).astype(str), cc=cur_column() ) )

Out[12]:

	x	y
	<object>	<object>
0	x 0.7739296633011361	y 0.05878489331508395
1	x 0.6233885082054422	y 0.027460112154048803
2	x 0.005457753705443728	y 0.03921965587769912
3	x 0.45944873370090106	y 0.034711398724083777
4	x 0.10502100613889181	y 0.04504004423820979
5	x 0.3479600358358678	y 0.9804449881028017

In [13]:

# or you can use x.name, since x is a Series
df >> mutate(across(
    [f.x, f.y], 
    lambda x: [x.name + ' '] * x.shape[0] + (x**2).astype(str)
))

# or you can use x.name, since x is a Series df >> mutate(across( [f.x, f.y], lambda x: [x.name + ' '] * x.shape[0] + (x**2).astype(str) ))

Out[13]:

	g	x	y
	<object>	<object>	<object>
0	a	x 0.7739296633011361	y 0.05878489331508395
1	b	x 0.6233885082054422	y 0.027460112154048803
2	b	x 0.005457753705443728	y 0.03921965587769912
3	c	x 0.45944873370090106	y 0.034711398724083777
4	c	x 0.10502100613889181	y 0.04504004423820979
5	c	x 0.3479600358358678	y 0.9804449881028017