Pandas: adding several columns to a dataframe in a single line

user_11133739 0
Viewed 297

I come from R background & am wondering if there's a single line code to add several new columns to an existing dataframe in Pandas just like dplyr. If have this code:

import pandas as pd

df = pd.DataFrame({'a': range(1, 11)})

df['b'] = range(11, 21)
df['c'] = range(21, 31)
df['d'] = range(31, 40)
df['e'] = range(41, 50)

Is there a way to make all columns addition into df in one line? An example of what I want in R would be:

library(dplyr)

df <- data.frame('a' = 1:10)

df <- df %>% mutate(b = 11:20, c = 21:30, d = 31:40, e = 41:50)
pandas dataframe python
5 Answers

There is assign:

df.assign(b=range(11,21), c=range(21,31), d=range(31,41))

Things are even easier when you have a dictionary:

# assume you get this from somewhere else
val_dict = {'b': range(11,21), 'c':range(21,31)}

df.assign(**val_dict)

Note the second approach is expected when b is not a possible choice for keyword arguments, for example, having spaces 'a b'.

user_42384080

As others have noted, you could build them all in the original construction of the dataframe, but if you needed to add multiple columns at a later point, you can add each through multiple declaration:

df['b'], df['c'], df['d'], df['e'] = range(11, 21), range(21,31), range(31,41), range(41,51)
user_113125370 
df = pd.DataFrame( {c: range(x, y) for c,x,y in [(chr(97+x), x*10+1, x*10+11) for x in range(5)]})

>>> df
    a   b   c   d   e
0   1  11  21  31  41
1   2  12  22  32  42
2   3  13  23  33  43
3   4  14  24  34  44
4   5  15  25  35  45
5   6  16  26  36  46
6   7  17  27  37  47
7   8  18  28  38  48
8   9  19  29  39  49
9  10  20  30  40  50

Or to add to an existing dataframe:

df = pd.DataFrame({'a': range(1,11)})
df = pd.concat([df, pd.DataFrame( {c: range(x, y) for c,x,y in [(chr(97+x), x*10+1, x*10+11) for x in range(1, 5)]})], axis=1)
user_24135480

Check out this:

>>> from datar.all import f, tibble, mutate
>>> df = tibble(a = f[1:10])
>>> df >> mutate(b = f[11:20], c = f[21:30], d = f[31:40], e = f[41:50])
        a       b       c       d       e
  <int64> <int64> <int64> <int64> <int64>
0       1      11      21      31      41
1       2      12      22      32      42
2       3      13      23      33      43
3       4      14      24      34      44
4       5      15      25      35      45
5       6      16      26      36      46
6       7      17      27      37      47
7       8      18      28      38      48
8       9      19      29      39      49
9      10      20      30      40      50

I am the author of the datar package.

user_50881650

You just pass all the data and associated column name into pd.DataFrame just as you did with column 'a', separate with commas.

Like this:

df = pd.DataFrame({'a': range(1, 11), 'b' : range(11, 21)})
user_94307890
Related