Extracting features from time with pandas

Some events occur more often at certain times of the day – for example, fraudulent activity is more likely to occur during the night or early morning. Air pollutant concentration also changes with the time of the day, with peaks at rush hour when there are more vehicles on the streets. Therefore, deriving time features is extremely useful. In this recipe, we will extract different time parts of a datetime variable by utilizing pandas and NumPy.

Getting ready

The following are some of the features that we can extract offtheshelf using pandas:

• pandas.Series.dt.hour
• pandas.Series.dt.minute
• pandas.Series.dt.second

How to do it...

To proceed with this recipe, we must import the necessary libraries and create a toy dataset:

1. Let’s import pandas and numpy:

   import numpy as np
   import pandas as pd

2. Let’s create 20 datetime observations, beginning from 2019-03-05 at midnight followed by increments...