How to Find Probability Distributions in Python

In data science and machine learning, understanding the nature and structure of your data is as important as the modeling phase. One way we achieve this is by finding the Probability Distribution of the variables in our data.

This tutorial will walk through how to compute probability distributions using Python. Python offers a plethora of libraries for various tasks but for this tutorial, we will be using Pandas, NumPy, and Matplotlib.

Step 1: Import the Required Libraries

Our first step is to import the libraries that we will be working with. These libraries provide us with the necessary functions for our computations. Here is the respective code:

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

Step 2: Load Your Dataset

To demonstrate how to calculate a probability distribution, we will need a dataset. Here, we load a dataset using the Pandas read_csv() function. You can use any dataset of choice:

1	data = pd.read_csv('data.csv')

You can use this sample data:

Assume the dataset, data.csv holds information on the ages of a population. We will find the probability distribution of the ages.

Step 3: Data Visualization

A good way to understand the distribution of your data is by visualizing it using a histogram. Here is a simple way to do it in Python:

1 2	plt.hist(data["age"], bins=20, density=True) plt.show()

Step 4: Calculation of the Probability Distribution

Now, let us calculate the actual probability distribution. We will calculate the probability of each unique value in our ‘age’ column and create a new DataFrame to store these probabilities.

age_counts = data['age'].value_counts()

total_ages = len(data['age'])

prob_distribution = pd.DataFrame(age_counts / total_ages)

Step 5: Visualization of the Probability Distribution

We can then visualize our calculated probability distribution as shown in the following code block:

1 2	plt.bar(prob_distribution.index, prob_distribution['age']) plt.show()

Full Python code

import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

data = pd.read_csv('data.csv')

plt.hist(data["age"], bins=20, density=True)

plt.show()