Pandas: Convert timezone-aware DateTimeIndex to naive timestamp

avatar
Borislav Hadzhiev

Last updated: Apr 12, 2024
4 min

banner

# Table of Contents

  1. Pandas: Convert timezone-aware DateTimeIndex to naive timestamp
  2. Using the tz_localize() method to get a naive local time
  3. Using the tz_convert() method to get a naive UTC time
  4. Pandas: Convert timezone-aware DateTimeIndex to naive timestamp using replace()
  5. Use apply() if you have a Series with multiple timezones
  6. The same approach can be used for a Pandas Timestamp

# Pandas: Convert timezone-aware DateTimeIndex to naive timestamp

To convert a timezone-aware DateTimeIndex to a naive timestamp, in a certain timezone:

  1. Use the tz_localize() method to remove the timezone information and get a naive local time.
  2. Or use the tz_convert() method to remove the timezone information and get the result in naive UTC time.

# Using the tz_localize() method to get a naive local time

The first example uses the tz_localize() method to convert a timezone-aware DateTimeIndex to a naive local time.

main.py
import pandas as pd

dt_index = pd.date_range(
    start='2023-04-11 13:30:00',
    periods=2,
    freq='H',
    tz='US/Pacific',
)

# DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')
print(dt_index)

dt_index = dt_index.tz_localize(None)

# DatetimeIndex(['2023-04-11 13:30:00', '2023-04-11 14:30:00'], dtype='datetime64[ns]', freq=None)
print(dt_index)
The code for this article is available on GitHub

Running the code sample produces the following output.

shell
DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')

DatetimeIndex(['2023-04-11 13:30:00', '2023-04-11 14:30:00'], dtype='datetime64[ns]', freq=None)

convert timezone aware datetimeindex to native timestamp

When the DatetimeIndex.tz_localize() method is called with a None value, it removes the time zone information and preserves the local time.

In other words, the result is a naive local time.

# Using the tz_convert() method to get a naive UTC time

If you need to convert a timezone-aware DateTimeIndex to naive UTC time, use the DateTimeIndex.tz_convert() method.

main.py
import pandas as pd

dt_index = pd.date_range(
    start='2023-04-11 13:30:00',
    periods=2,
    freq='H',
    tz='US/Pacific',
)

# DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')
print(dt_index)


dt_index = dt_index.tz_convert(None)

# DatetimeIndex(['2023-04-11 20:30:00', '2023-04-11 21:30:00'], dtype='datetime64[ns]', freq='H')
print(dt_index)
The code for this article is available on GitHub

Running the code sample produces the following output.

shell
DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')

DatetimeIndex(['2023-04-11 20:30:00', '2023-04-11 21:30:00'], dtype='datetime64[ns]', freq='H')

using tz convert method to get naive utc time

When the DateTimeIndex.tz_convert() method is called with a None value, it converts the datetime to UTC and removes the timezone information.

In other words, the result is a naive UTC time.

The DatetimeIndex.tz_convert() method is also often used to convert a timezone-aware Datetime array or index from one timezone to another.

main.py
import pandas as pd

dt_index = pd.date_range(
    start='2023-04-11 13:30:00',
    periods=2,
    freq='H',
    tz='US/Pacific',
)

# DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')
print(dt_index)


dt_index = dt_index.tz_convert('US/Central')

# DatetimeIndex(['2023-04-11 15:30:00-05:00', '2023-04-11 16:30:00-05:00'], dtype='datetime64[ns, US/Central]', freq='H')
print(dt_index)
The code for this article is available on GitHub

Running the code sample produces the following output.

shell
DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')

DatetimeIndex(['2023-04-11 15:30:00-05:00', '2023-04-11 16:30:00-05:00'], dtype='datetime64[ns, US/Central]', freq='H')

using tz convert to convert one timezone to another

The only argument we passed to the DateTimeIndex.tz_convert() method is the timezone.

The timestamps in the array/index get converted to the specified timezone.

As previously noted, if the tz argument is set to None, the method converts the datetime to UTC and removes the timezone information.

# Pandas: Convert timezone-aware DateTimeIndex to naive timestamp using replace()

You can also use the Timestamp.replace() method to convert a timezone-aware DateTimeIndex to a naive timestamp.

main.py
import pandas as pd

dt_index = pd.date_range(
    start='2023-04-11 13:30:00',
    periods=2,
    freq='H',
    tz='US/Pacific',
)

# DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')
print(dt_index)


dt_index = pd.DatetimeIndex(
    [tstamp.replace(tzinfo=None) for tstamp in dt_index]
)

# DatetimeIndex(['2023-04-11 13:30:00', '2023-04-11 14:30:00'], dtype='datetime64[ns]', freq=None)
print(dt_index)
The code for this article is available on GitHub

Running the code sample produces the following output.

shell
DatetimeIndex(['2023-04-11 13:30:00-07:00', '2023-04-11 14:30:00-07:00'], dtype='datetime64[ns, US/Pacific]', freq='H')

DatetimeIndex(['2023-04-11 13:30:00', '2023-04-11 14:30:00'], dtype='datetime64[ns]', freq=None)

convert timezone aware datetimeindex to naive timestamp using replace

We used a list comprehension to iterate over the DatetimeIndex.

List comprehensions are used to perform some operation for every element or select a subset of elements that meet a condition.

On each iteration, we use the Timestamp.replace() method to create a timezone unaware object from an aware object.

main.py
dt_index = pd.DatetimeIndex(
    [tstamp.replace(tzinfo=None) for tstamp in dt_index]
)

When the tzinfo argument is set to None, the method removes the timezone information.

# Use apply() if you have a Series with multiple timezones

You can use the Series.apply() method if you have a Series with multiple different timezones.

main.py
import pandas as pd

series = pd.Series([pd.Timestamp.now(tz='US/Pacific'),
                    pd.Timestamp.now(tz='Europe/Amsterdam'),
                    pd.Timestamp.now(tz='Europe/Brussels')])

series = series.apply(lambda x: x.tz_localize(None))

# 0   2023-08-22 06:48:19.822881
# 1   2023-08-22 15:48:19.824277
# 2   2023-08-22 15:48:19.824789
# dtype: datetime64[ns]
print(series)

use apply for series with multiple timezones

The code for this article is available on GitHub

The Series.apply() method invokes a function on each value of the Series.

We called the tz_localize() method on each timestamp.

If you need to get the results in UTC, use the tz_convert method instead.

main.py
import pandas as pd

series = pd.Series([pd.Timestamp.now(tz='US/Pacific'),
                    pd.Timestamp.now(tz='Europe/Amsterdam'),
                    pd.Timestamp.now(tz='Europe/Brussels')])

series = series.apply(lambda x: x.tz_convert(None))

# 0   2023-08-22 13:51:12.547237
# 1   2023-08-22 13:51:12.547991
# 2   2023-08-22 13:51:12.548291
# dtype: datetime64[ns]
print(series)

use tz convert to get result in utc

# The same approach can be used for a Pandas Timestamp

The same approach can be used if you have a Pandas Timestamp object.

For example, if you need to convert a timezone-aware Timestamp to a naive timestamp and get the result in local time, use tz_localize().

main.py
import pandas as pd

tstamp = pd.Timestamp.now(tz='US/Pacific')

# 2023-08-22 06:42:03.820451-07:00
print(tstamp)

naive_tstamp = tstamp.tz_localize(None)

# 2023-08-22 06:42:03.820451
print(naive_tstamp)
The code for this article is available on GitHub

If you need to get the result in UTC, use the tz_convert() method instead.

main.py
import pandas as pd

tstamp = pd.Timestamp.now(tz='US/Pacific')

# 2023-08-22 06:42:03.820451-07:00
print(tstamp)

naive_tstamp = tstamp.tz_convert(None)

# 2023-08-22 13:42:03.820451
print(naive_tstamp)

This is equivalent to calling the utcnow() method before calling tz_localize().

main.py
import pandas as pd

tstamp = pd.Timestamp.now(tz='US/Pacific')

# 2023-08-22 06:42:03.820451-07:00
print(tstamp)

naive_tstamp = tstamp.utcnow().tz_localize(None)

# 2023-08-22 13:42:03.820451
print(naive_tstamp)

Or calling the utcnow() method before calling tz_convert().

main.py
import pandas as pd

tstamp = pd.Timestamp.now(tz='US/Pacific')

# 2023-08-22 06:42:03.820451-07:00
print(tstamp)

naive_tstamp = tstamp.utcnow().tz_convert(None)

# 2023-08-22 13:42:03.820451
print(naive_tstamp)

# Additional Resources

You can learn more about the related topics by checking out the following tutorials:

I wrote a book in which I share everything I know about how to become a better, more efficient programmer.
book cover
You can use the search field on my Home Page to filter through all of my articles.
About
Contacts
Policy
Terms & Conditions
TwitterGitHubLinkedin

Copyright © 2024 Borislav Hadzhiev

0
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........