Python Feature Engineering Cookbook

How to do it...

To proceed with the recipe, let's import the libraries and create a toy dataset:

1. Let's begin by importing pandas, numpy, and datetime:

import datetimeimport numpy as npimport pandas as pd

2. Let's create 2 datetime variables with 20 values each, in which values start from 2019-05-03 and increase in intervals of 1 hour or 1 month, respectively. Then, let's capture the variables in a dataframe, add column names, and display the top rows:

rng_hr = pd.date_range('2019-03-05', periods=20, freq='H')rng_month = pd.date_range('2019-03-05', periods=20, freq='M')df = pd.DataFrame({'date1': rng_hr, 'date2': rng_month})  df.head()

We can see the first five rows of the created variables in the following output:

3. Let's capture the difference ...