Finally, let's move to the last method of this section called .corr(). It helps out a lot to find the relationship between numerical data. Imagine that you have a dataset on houses:

Let's examine the output of the data.corr() in our case:

So, let's do it step by step: You have vertical and horizontal values; each pair overlaps. In each overlap, we can receive a value from -1 to 1.

• 1 means that two values depend on each other in a directly proportional way (if one value increases, the other increases too);
• -1 means that two values depend on each other in an inversely proportional way (if one value increases, the other decreases);
• 0 means that the two dependent values aren't proportional.

Note

If the dataset contains non-numeric columns, such as in the cars.csv dataset used in the task, you should set the argument numeric_only=True to compute the correlation using only the numeric columns.