An Overview of Python

INTRO

I recently joined an amazing coding bootcamp focused on data science leveraging Python, a computer language. Though I do have a degree in the science of Biochemistry, Data Science and Python are completely different from any scientific academic endeavor I’ve undertaken before.

I like to think of Data Science with python as as complex as learning a new language while leveraging mathematics in tandem to learning how to think in a manner that optimizes your work output. I know it sounds scary and overwhelming, and believe me, I thought that too at the beginning. However, once we dissect all the tools and communities at our disposal working with python becomes an amazing experience applicable to many disciplines and industries.

OVERVIEW OF PYTHON AND OPEN SOURCE COMMUNITY

Python was created by a Dutch software engineer named, Guido van Rossum in the late ’80s and was deployed by 1991. Rossum’s goal with this monumental “side project” was to develop a coding language that would have easier code readability, while letting the coder express concepts in a more concise syntax, and therefore fewer lines of code as well. The superiority of Python was undeniable and it soon replaced its predecessor programming language, ABC, which Rossum had also created. Rossum achieved his goal of making the language simplified and concise that even 30 years later it is still considered as one of the best languages for a new student of data science to being their journey with.

As time went on the software engineering and data science industries took a very unique approach to intellectual property. Companies and developers openly shared the code for their projects and anyone with a connection to the internet was able to not only look at them but participate in enhancing each project. Python’s functionality increased as people submitted patches that would allow for more complex features such as functional style of programming with tools such as lambda, map filter, and reduce.

LIBRARIES AS TOOLS

This is where Python goes into overdrive. As I just mentioned, throughout the years Python has been supplemented with additional patches and packages of code developed by independent parties and is accessible to everyone. These additions or supplements are referred to as Libraries and can be imported into any project. Think of Libraries like applications on your phone. You interact with it the same way but it allows you to take advantage of your phone’s capabilities in completely different ways.

Let us explore some popular libraries and see how they help us maximize our capabilities while interacting with data in any given project:

• Pandas (“import pandas as pd”): Pandas is one the most fundamental libraries used when analyzing data. If you’re working with any sort of data, you’re going to use Pandas. With pandas, you can load, prepare, manipulate, model, and analyze data. You can join data frames, merge them, merge only the left, merge only the right, you can reshape data, you can work with different databases. The array of possibilities to analyze data is endless.

• NumPy (“import numpy as np”): Numpy is essential for working with numbers, calculations, data science, and machine learning, its name says it all — Numerical Python! An important characteristic of NumPy is that it can iterate over complex and lengthy equations in fractions of a second, making it exponentially faster than the alternative of a simple Python list or loop.

• Matplotlib (“import matplotlib.pyplot as plt”): Matplotlib is built-in for Pandas and it’s a library dedicated to the visualization of data. You can also look at images and manipulate images. It gives us ample control over visualizing our data and it has other libraries dedicated to it, such as Seaborn which makes data visualization even easier and more detailed.

APPLICABILITY

I first heard of python in a presentation at Goldman Sachs about the future of technology in finance. In that particular scenario, Python allowed data scientists to calculate complex economic modules through machine learning and AI. However, python’s functionality and applicability can be leveraged in just about any industry, not just finance.

In 2017, The Economist published an article titled ‘The world’s most valuable resource is no longer oil, but data.’. Data and its analyses allow just about every business to maximize its strengths and cut its losses. Megacorporations and medical research teams can analyze data from millions of customers and patients with in-depth visualizations leveraging Python along with Seaborn and SQL databases. Conveniently enough, it can also be used in small or developing businesses given its simplicity, concise syntax, and ample resources. Oh, and also, it’s free!

CONCLUSION

Today Python is on version 3.9.0 and its usefulness, tools, and communities only keep on growing exponentially. With the advent and proliferation of machine learning and AI into every industry, one can only be certain that Python’s applicability has only begun to be tapped into.