Machine Learning with Python Cookbook, 2nd Edition

Chapter 7. Handling Dates and Times

7.0 Introduction

Dates and times (datetimes), such as the time of a particular sale or the date of a public health statistic, are frequently encountered during preprocessing for machine learning. Longitudinal data (or time series data) is data that’s collected repeatedly for the same variables over points in time. In this chapter, we will build a toolbox of strategies for handling time series data, including tackling time zones and creating lagged time features. Specifically, we will focus on the time series tools in the pandas library, which centralizes the functionality of many other general libraries such as datetime.

7.1 Converting Strings to Dates

Problem

Given a vector of strings representing dates and times, you want to transform them into time series data.

Solution

Use pandas’ to_datetime with the format of the date and/or time specified in the format parameter:

# Load libraries
import numpy as np
import pandas as pd

# Create strings
date_strings = np.array(['03-04-2005 11:35 PM',
                         '23-05-2010 12:01 AM',
                         '04-09-2009 09:09 PM'])

# Convert to datetimes
[pd.to_datetime(date, format='%d-%m-%Y %I:%M %p') for date in date_strings]

[Timestamp('2005-04-03 23:35:00'),
 Timestamp('2010-05-23 00:01:00'),
 Timestamp('2009-09-04 21:09:00')]

We might also want to add an argument to the errors parameter to handle problems:

# Convert to datetimes
[pd.to_datetime(date, format="%d-%m-%Y %I:%M %p", errors="coerce")
for date in date_strings]

[Timestamp('2005-04-03 ...