Issue #416 (April 14, 2020) (39)

Open Source Virtual Backgrounds With Python and OpenCV With so much of the world stuck at home thanks to the COVID-19 pandemic, more and more workplaces are using video conferencing solutions—such as Zoom—to bring work to your living room. Or kitchen. Or bedroom. Or bathroom. OK… hopefully not that last one! Why not spice up your video conferencing with some awesome, homemade virtual backgrounds...

Inheritance and Composition: A Python OOP Guide Learn about inheritance and composition in Python. You’ll improve your object-oriented programming (OOP) skills by understanding how to use inheritance and composition and how to leverage them in their design. REAL PYTHON course

Stop Naming Your Python Modules utils “Do not use utils as a name for your Python module neither put it into a class name. Try to be more specific about the role of code – e.g. create a place for validators, services or factories. If the role criterion doesn’t help and you really dealing with utils, try grouping code by its theme.” SEBASTIAN BUCZYŃSKI...

From Chunking to Parallelism: Faster Pandas With Dask In a recent article, Itamar Turner-Trauring discussed how to read large datasets with Pandas using a chunking technique to reduce memory overhead. Chunking has another advantage: it enables parallelism, which can dramatically speed up processing time. Learn why and how to enable parallelism for your Pandas processing code using the Dask library. ITAMAR TURNER-TRAURING

Data Science: Reality Doesn’t Meet Expectations “Seven common ways a data science role may not meet your expectations through tens of data scientist interviews and anecdotes from popular media.” Also see the related discussion on Hacker News DAN FRIEDMAN opinion

How to Provide Test Fixtures for Django Models in pytest In this step-by-step tutorial, you’ll learn how to use fixtures to simplify your testing in pytest. If you use Django, pytest fixtures can help you write tests that are painless to update and maintain even as you make changes to your models. REAL PYTHON

Can You Retrieve Elements Silently Consumed by zip() From Generators of Unequal Length? If you pass two generators to zip(), then the iterator returned by zip() stops whenever the shorter of the two generators runs out of values. If the generator passed to zip() in the first position is larger than the second, then there is an “extra” value consumed from the larger generator....

How Can I Simplify Repetitive if/elif Statements? Comparing a single value against a number of conditions can result in long if/elif blocks. Is there a short and clean way to handle these comparisons? STACK OVERFLOW

CuPy Accelerates NumPy on the GPU? Hold My Cider, Here’s Clojure! NumPy is fast, but it doesn’t play well with GPUs. For that, you can use CuPy, a drop-in NumPy replacement for GPUs developed by Nvidia. Now, what if I told you to crunch those numbers in Clojure on a GPU instead? Surely a language that compiles to Java couldn’t beat C++ optimized for...

3 Ways to Test S3 in Python Testing an application’s interaction with a third-party service can be difficult and frustrating. But sometimes a lot of business logic is wrapped up in those interactions, so testing them is important for building trust in the application. This article compares and contrasts three options for testing Amazon S3 integration using the pytest framework. SANJAY SIDDHANTI

Neural Network Showdown: TensorFlow Vs PyTorch TensorFlow and PyTorch are the two Python libraries that have accelerated the use of neural networks. This post goes over their pros and cons, so you can decide which to use in your next data science project. Check out ActiveState’s blog → ACTIVESTATE sponsor

Quick Domain-Specific Languages in Python With textX Domain-specific languages (DSL) bring the power of high-level programming to domain experts (read: non-programmers). In this article, Federico Tomassetti walks you through the process of creating a DSL with the textX Python library. As a bonus, you’ll also learn how to create syntax highlighting extension for VS Code that understands your language. FEDERICO TOMASSETTI

Visualizing Decision Trees With Python “Decision trees are a popular supervised learning method for a variety of reasons. Benefits of decision trees include that they can be used for both regression and classification, they don’t require feature scaling, and they are relatively easy to interpret as you can visualize decision trees. This is not only a powerful way to understand your model, but also...

John Conway, Inventor of the Game of Life, Has Died of COVID-19 John Horton Conway was one of the most influential mathematicians of the 20th and 21st centuries. His “Game of Life” spawned countless implementation—nearly becoming a rite of passage for programmers in any language. Sadly, the world lost this brilliant mind on April 11 due to complications from COVID-19. Renowned mathematician Terry Tao...

Modeling Stock Portfolios With Python After reading an article about comparing a stock portfolio to the S&P 500, Matt Grierson wanted to go deeper and model a portfolio’s performance by analyzing buys and sells from a CSV file and calculating the rate of return relative to an index across any specified timeframe. This turned out to be more involved than Matt first thought it...

Adventures in Manipulating Python ASTs George Ho proposed a change to PyMC4’s model specification API that he thought would help make the user experience more straightforward. Although those changes weren’t ultimately accepted by his fellow PyMC4 core developers, the process led him into some interesting explorations into Python’s abstract syntax tree. GEORGE HO

RPP Episode #4: Learning Python Through Errors An interview with Martin Breuss on getting started with Django, and how to learn Python through errors, and how errors really are your friends. Martin talks about his work with Coding Nomads, and teaching Python around the world. He also provides some tips on debugging and writing good questions. REAL PYTHON podcast

Combining Data in Pandas In this step-by-step tutorial, you’ll learn three techniques for combining data in Pandas: merge, join, and concat. Combining Series and DataFrame objects in Pandas is a powerful way to gain new insights into your data. REAL PYTHON

Learn Python for Data Science in 4 Weeks Learn the foundational Python programming and statistics skills needed for a job in data science in as little as 4 weeks. Work 1:1 with a data science mentor to master the skills needed to get started in your journey to a data science role. Enroll in Springboard’s data science career track prep course today. SPRINGBOARD sponsor

Scout APM for Python Check out Scout’s developer-friendly application performance monitoring solution for Python. Scout continually tracks down N+1 database queries, sources of memory bloat, performance abnormalities, and more. Get back to coding with Scout → SCOUT APM sponsor