How to create multiple additional columns from dataframe and add to the same dataframe

Question

Trying to parse 'date' column into 'month', 'day', 'hour' and 'minute' and then add them as separate columns to the same dataframe:

import pandas as pd
d = {'date':[pd.Timestamp('2019-03-01 00:05:01'),
             pd.Timestamp('2019-04-02 07:11:00'),
             pd.Timestamp('2019-05-03 10:25:00')], 
     'foo': ['abc','def','jhk']
}
df1 = pd.DataFrame(d)

    date                foo
0   2019-03-01 00:05:01 abc
1   2019-04-02 07:11:00 def
2   2019-05-03 10:25:00 jhk

After extracting 'times':

times = df1['date'].apply(lambda date: (date.month, date.day, date.hour, date.minute))

I try to add them to the dataframe as separate columns:

df1['month'], df1['day'], df1['hour'], df1['minute'] = times

Which results in error:

 ValueError                                Traceback (most recent call last)
 <ipython-input-21-171174d71b13> in <module>
  ----> 1 df1['month'], df1['day'], df1['hour'], df1['minute'] = times

 ValueError: not enough values to unpack (expected 4, got 3)

How to add 'times' as separate columns?

yatu · Accepted Answer · 2020-01-14 10:57:01Z

1

Looks like you want

df1['month'], df1['day'], df1['hour'], df1['minute'] =  (df1.date.dt.month, df1.date.dt.day, 
                                                        df1.date.dt.hour, df1.date.dt.minute)

print(df1)

date  foo  month  day  hour  minute
0 2019-03-01 00:05:01  abc      3    1     0       5
1 2019-04-02 07:11:00  def      4    2     7      11
2 2019-05-03 10:25:00  jhk      5    3    10      25

answered Jan 14, 2020 at 10:57

yatu

88.6k12 gold badges93 silver badges148 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

dokondr Over a year ago

When dataframe 'assign' method is better to use then dot notation in your solution?

Shijith · Accepted Answer · 2020-01-14 11:04:20Z

1

Alternatively, use pd.assign:

df1.assign(month=df1["date"].dt.month, day=df1["date"].dt.day, hour=df1["date"].dt.hour, minutes=df1["date"].dt.minute)

Output:

                  date  foo  month  day  hour   minutes
0   2019-03-01 00:05:01 abc     3     1    0          5
1   2019-04-02 07:11:00 def     4     2    7         11
2   2019-05-03 10:25:00 jhk     5     3   10         25

edited Jan 14, 2020 at 11:04

Shijith

4,9322 gold badges23 silver badges38 bronze badges

answered Jan 14, 2020 at 10:58

Fourier

3,0033 gold badges29 silver badges40 bronze badges

2 Comments

dokondr Over a year ago

When 'pd.assign' is better to use then dot notation?

Fourier Over a year ago

@dokondr There is actually no pd.assign, assign is a function of the DataFrame object: pandas.pydata.org/pandas-docs/stable/reference/api/…

Collectives™ on Stack Overflow

How to create multiple additional columns from dataframe and add to the same dataframe

2 Answers 2

1 Comment

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related