Reading date and time data in Pandas
Working with Dates and Times in Python
Max Shron
Data Scientist and Author
A simple Pandas example
# Load Pandas import pandas as pd# Import W20529's rides in Q4 2017 rides = pd.read_csv('capital-onebike.csv')
A simple Pandas example
# See our data
print(rides.head(3))
Start date End date Start station \
0 2017-10-01 15:23:25 2017-10-01 15:26:26 Glebe Rd & 11th St N
1 2017-10-01 15:42:57 2017-10-01 17:49:59 George Mason Dr & Wilson Blvd
2 2017-10-02 06:37:10 2017-10-02 06:42:53 George Mason Dr & Wilson Blvd
End station Bike number Member type
0 George Mason Dr & Wilson Blvd W20529 Member
1 George Mason Dr & Wilson Blvd W20529 Casual
2 Ballston Metro / N Stuart & 9th St N W20529 Member
A simple Pandas example
rides['Start date']
0 2017-10-01 15:23:25
1 2017-10-01 15:42:57
...
Name: Start date, Length: 290, dtype: object
rides.iloc[2]
Start date 2017-10-02 06:37:10
End date 2017-10-02 06:42:53
...
Name: 1, dtype: object
Loading datetimes with parse_dates
# Import W20529's rides in Q4 2017 rides = pd.read_csv('capital-onebike.csv', parse_dates = ['Start date', 'End date'])# Or: rides['Start date'] = pd.to_datetime(rides['Start date'], format = "%Y-%m-%d %H:%M:%S")
Loading datetimes with parse_dates
# Select Start date for row 2
rides['Start date'].iloc[2]
Timestamp('2017-10-02 06:37:10')
Timezone-aware arithmetic
# Create a duration column rides['Duration'] = rides['End date'] - rides['Start date']# Print the first 5 rows print(rides['Duration'].head(5))
0 0 days 00:03:01
1 0 days 02:07:02
2 0 days 00:05:43
3 0 days 00:21:18
4 0 days 00:21:17
Name: Duration, dtype: timedelta64[ns]
Loading datetimes with parse_dates
rides['Duration']\
.dt.total_seconds()\
.head(5)
0 181.0
1 7622.0
2 343.0
3 1278.0
4 1277.0
Name: Duration, dtype: float64
Reading date and time data in Pandas
Working with Dates and Times in Python
