Plotting histogram

To get a better understanding of the data, we plot histogram for each numerical attribute. It shows us the number of instances that lie between a particular range.

Let's plot a histogram for an arbitrarily chosen dataset-

enter image description here

So, on seeing the above histogram, we can conclude that-

There are 100 instances in the dataset whose value lie between 0 and 1.
There are 40 instances in the dataset whose value lie between 1 and 2.
There are 20 instances in the dataset whose value lie between 2 and 3.
There are 60 instances in the dataset whose value lie between 3 and 4.
There are 80 instances in the dataset whose value lie between 4 and 5.

We do this generally for numerical attributes as we can see the count of instances belonging to each category of a categorical attribute by value_counts() method of the DataFrame object which we have done before, because it gives us exact figures of the count.

We plot a histogram by calling the hist() method of the DataFrame object. It calls the hist() method of matplotlib.pyplot internally , on each attribute in the DataFrame, resulting in one histogram per column. Hence, we have to first import matplotlib.pyplot to make it work.

Here, matplotlib is a module and pyplot is a sub-module of it. Most of the matplotlib utilities lie under pyplot. It is generally imported under the plt alias.

Previous Index Next

End-to-End ML Project- Beginner friendly

Plotting histogram

XP

Loading comments...