Datetime Methods Reference¶

Complete reference for all datetime methods and properties available through the pandas dt accessor.

Date Component Properties¶

```python import pandas as pd

dates = pd.Series(pd.date_range('2024-03-15', periods=3))

print(dates.dt.year) # 2024, 2024, 2024 print(dates.dt.month) # 3, 3, 3 print(dates.dt.day) # 15, 16, 17 print(dates.dt.quarter) # 1, 1, 1 print(dates.dt.dayofweek) # 4, 5, 6 (Fri, Sat, Sun) print(dates.dt.dayofyear) # 75, 76, 77 print(dates.dt.days_in_month) # 31, 31, 31 ```

Time Component Properties¶

```python times = pd.Series(pd.to_datetime(['2024-01-15 14:30:45.123456']))

print(times.dt.hour) # 14 print(times.dt.minute) # 30 print(times.dt.second) # 45 print(times.dt.microsecond) # 123456 print(times.dt.nanosecond) # 0 ```

Date/Time Extraction Properties¶

```python times = pd.Series(pd.to_datetime(['2024-01-15 14:30:45']))

print(times.dt.date) # 2024-01-15 print(times.dt.time) # 14:30:45 ```

Name Methods¶

```python dates = pd.Series(pd.date_range('2024-01-15', periods=3))

print(dates.dt.day_name())

Monday, Tuesday, Wednesday¶

print(dates.dt.month_name())

January, January, January¶

With locale (if supported)¶

print(dates.dt.day_name(locale='de_DE'))

Montag, Dienstag, Mittwoch¶

```

Boolean Properties¶

```python dates = pd.Series(pd.to_datetime([ '2024-01-01', # Year start, Q1 start, month start '2024-03-31', # Q1 end, month end '2024-06-15', # Mid-quarter '2024-12-31' # Year end, Q4 end, month end ]))

print(dates.dt.is_month_start) # True, False, False, False print(dates.dt.is_month_end) # False, True, False, True print(dates.dt.is_quarter_start) # True, False, False, False print(dates.dt.is_quarter_end) # False, True, False, True print(dates.dt.is_year_start) # True, False, False, False print(dates.dt.is_year_end) # False, False, False, True print(dates.dt.is_leap_year) # True, True, True, True ```

Week Methods¶

```python dates = pd.Series(pd.date_range('2024-01-01', periods=5))

iso = dates.dt.isocalendar() print(iso)

year week day¶

0 2024 1 1¶

1 2024 1 2¶

...¶

Access individual components¶

print(dates.dt.isocalendar().week) # Week numbers ```

Formatting Methods¶

strftime(format)¶

Format datetime as string using format codes.

```python dates = pd.Series(pd.date_range('2024-01-15', periods=3))

print(dates.dt.strftime('%Y-%m-%d'))

2024-01-15, 2024-01-16, 2024-01-17¶

print(dates.dt.strftime('%d/%m/%Y'))

15/01/2024, 16/01/2024, 17/01/2024¶

print(dates.dt.strftime('%B %d, %Y'))

January 15, 2024, January 16, 2024, January 17, 2024¶

print(dates.dt.strftime('%A, %B %d'))

Monday, January 15, Tuesday, January 16, Wednesday, January 17¶

```

Format Code Reference¶

Rounding Methods¶

```python times = pd.Series(pd.to_datetime([ '2024-01-15 09:37:45', '2024-01-15 14:22:30', '2024-01-15 18:45:15' ]))

Round to hour¶

print(times.dt.round('H'))

10:00, 14:00, 19:00¶

Floor to hour¶

print(times.dt.floor('H'))

09:00, 14:00, 18:00¶

Ceil to hour¶

print(times.dt.ceil('H'))

10:00, 15:00, 19:00¶

Round to 15 minutes¶

print(times.dt.round('15T'))

09:30, 14:30, 18:45¶

Round to day¶

print(times.dt.round('D'))

2024-01-16, 2024-01-15, 2024-01-16¶

```

Frequency Strings¶

Timezone Methods¶

```python

Create naive datetime¶

dates = pd.Series(pd.date_range('2024-01-15 12:00', periods=3, freq='H'))

Localize to UTC¶

dates_utc = dates.dt.tz_localize('UTC') print(dates_utc)

2024-01-15 12:00:00+00:00, ...¶

Convert to US Eastern¶

dates_eastern = dates_utc.dt.tz_convert('US/Eastern') print(dates_eastern)

2024-01-15 07:00:00-05:00, ...¶

Get timezone¶

print(dates_utc.dt.tz) # UTC print(dates_eastern.dt.tz) # US/Eastern ```

Common Timezone Strings¶

Normalization Method¶

normalize()¶

Set time component to midnight.

```python times = pd.Series(pd.to_datetime([ '2024-01-15 09:30:00', '2024-01-15 14:45:00', '2024-01-16 23:59:59' ]))

print(times.dt.normalize())

2024-01-15, 2024-01-15, 2024-01-16 (all at 00:00:00)¶

```

Timedelta Properties¶

For Series with timedelta64 dtype:

```python

Create timedelta Series¶

durations = pd.Series([ pd.Timedelta(days=1, hours=2, minutes=30), pd.Timedelta(hours=5, minutes=45), pd.Timedelta(days=2, seconds=3600) ])

print(durations.dt.days) # 1, 0, 2 print(durations.dt.total_seconds()) # 95400.0, 20700.0, 176400.0 print(durations.dt.components)

days hours minutes seconds milliseconds microseconds nanoseconds¶

0 1 2 30 0 0 0 0¶

1 0 5 45 0 0 0 0¶

2 2 1 0 0 0 0 0¶

```

Period Properties¶

For Series with Period dtype:

```python

Create period Series¶

periods = pd.Series(pd.period_range('2024-01', periods=3, freq='M'))

print(periods.dt.start_time) # Start of each month print(periods.dt.end_time) # End of each month print(periods.dt.freq) # ```

Method Summary¶

Exercises¶

Exercise 1. Create a Series of 12 monthly timestamps for the year 2024. Use .dt.month_name() to get the month names and .dt.is_quarter_start to identify which months are quarter starts.

import pandas as pd

months = pd.Series(pd.date_range('2024-01-01', periods=12, freq='MS'))
print(months.dt.month_name())
print("Quarter starts:")
print(months[months.dt.is_quarter_start])

Exercise 2. Given a Series of timestamps in UTC, use .dt.tz_localize('UTC') and then .dt.tz_convert('US/Eastern') to convert them to Eastern time. Print the resulting timestamps and confirm the timezone offset.

import pandas as pd

timestamps = pd.Series(pd.to_datetime([
    '2024-06-15 18:00:00',
    '2024-12-15 18:00:00'
]))
utc = timestamps.dt.tz_localize('UTC')
eastern = utc.dt.tz_convert('US/Eastern')
print(eastern)

Exercise 3. Create a Series of timestamps with different times of day. Use .dt.strftime('%Y/%m/%d %I:%M %p') to format them in 12-hour format with AM/PM. Then use .dt.normalize() to strip the time component and keep only dates.

import pandas as pd

ts = pd.Series(pd.to_datetime([
    '2024-01-15 09:30:00',
    '2024-01-15 14:45:00',
    '2024-01-15 21:00:00'
]))
formatted = ts.dt.strftime('%Y/%m/%d %I:%M %p')
print("Formatted:\n", formatted)

normalized = ts.dt.normalize()
print("Normalized:\n", normalized)