Types consistency

One of the first steps of analyzing received data is checking the values types. If we are talking about column with age, then we expect to have there integer type; or column with salaries should be either integer or float.

Remember, to get the columns types in pandas, you should use the .dtypes attribute. Execute the code below to find out values types.


              1234567
            
# Importing the library
import pandas as pd

# Reading the file
df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/f2947b09-5f0d-4ad9-992f-ec0b87cd4b3f/data.csv')
# Output values types
print(df.dtypes)

So many types... How not to get confused here? Let's see what columns have object type. To do it, we are going to use the same attribute and within square brackets set the condition. Since we received a Series object, column names will be indexes of this Series, so for convenient output we will output indexes only.


              1234567
            
# Importing the library
import pandas as pd

# Reading the file
df = pd.read_csv('https://codefinity-content-media.s3.eu-west-1.amazonaws.com/f2947b09-5f0d-4ad9-992f-ec0b87cd4b3f/data.csv')
# Output only columns with 'object' type
print(df.dtypes[df.dtypes == object].index)

Here is the first problem. Columns 'totinch', 'morgh', 'valueh', 'grosrth', 'omphtotinch' should be considered as numerical, taking into account their specifics.

Column	Description
`'TOTINCH'`	Total Household Income
`'MORGH'`	Presence of Mortgage
`'VALUEH'`	Value of Dwelling
`'GROSRTH'`	Monthly Gross Rent
`'OMPH'`	Owner's Major Payments (Monthly)

Let's find out why these columns were considered as object.

Everything was clear?

Section 1. Chapter 2