In this exercise, you are going to analyze a new dataset in terms of linearity, correlation, and the statistical significance of the means of certain categories and whether some variables follow a normal distribution.

The dataset is the bike sharing dataset available from the UCI repository. This dataset has daily and and hourly data. We are going to work on the day.csv file which has 731 samples and 16 different variables. 

And we are going to focus on the following five attributes:

• Season: Season (1:spring, 2:summer, 3:fall, 4:winter)
• Temp: Normalized temperature in Celsius. 
• Hum: Normalized humidity. 
• Wind speed: Normalized wind speed. 
• CNT: Count of total rental bikes. 

You can load the dataset with:

import pandas as pd
df = pd.read_csv('day.csv')

And remove the non essential columns with:

df = df[['season', 'temp','hum','windspeed','cnt']]