Wrong Convertation?

Well, we have found out that the initial data contains dates in the format dd-mm-yyyy. The datetime module reads dates in the yyyy-mm-dd format, which is fine. However, how will datetime convert the date 2010-05-02? Since we are familiar with the dataset, we know that it represents February 5, 2010. But datetime doesn't know this and may consider this date as May 2, 2010. Let's check if this has happened in our data.


              123456789101112131415
            
# Loading the library
import pandas as pd

# Reading the data
df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/72be5dde-f3e6-4c40-8881-e1d97ae31287/shops_data3.csv')

# Displaying first four dates before converting
print(df['Date'].head(4))

# Change column type
df['Date'] = pd.to_datetime(df['Date'])

# Displaying first four dates and dtypes of dataframe
print(df['Date'].head(4))
print(df.dtypes)

Unfortunately, it has. Let's compare the first three dates in their initial and converted formats:

Initial date	Converted date
05-02-2010	2010-05-02
12-02-2010	2010-12-02
19-02-2010	2010-02-19

It is visible that for the first two dates, the left numbers were chosen as months, while for the third date, the left number was chosen as the day (the correct approach). Therefore, we need to make pandas understand the format of the dates we are working with.

Все було зрозуміло?

Секція 2. Розділ 4