Python In Depth
One of the features I love most about working in Jupyter Lab and Jupyter Notebooks is that it supports visual, interactive experiments with code and documenting what you’re doing in Markdown. With tools like Binder or CoCalc, it’s also possible to publish fully live Jupyter Notebooks with which programmers can interact. I haven’t availed myself … Read more
One of those questions that every Python developer wonders about at some point are those weird extra directories and files they’ll see crop up from time to time. Setting aside __init__.py for a moment, sometimes you’ll see directories you didn’t create: __pycache__, filled with files with a .pyc extension that have the same root filename … Read more
When I first took a look at Julia, I was wondering if it would be easy to learn and use it, coming from a background as a Python developer. I thought features like LISP-style macros might make it a difficult transition. It turns out it was easier than expected. Julia has many features that are … Read more
Today I spent some time taking a look at PyPy. I had expected it to be much faster than the “standard” version of Python, also known as CPython. However, I also wanted to account for speed differences when it comes to NumPy, a commonly used Python library written in C. PyPy is faster than CPython … Read more
When newcomers think of arrays, they may have an idea that this is implemented in Python as the built-in list type. However, as in other languages, Python lists and arrays are actually quite different. Even experienced Python developers often reach for the list type first when they need a sequence type, and for most cases, … Read more
The easiest way to compare Python dictionaries is simply to use the same equality operator you use on other Python types, “==”. This works great for most cases, but it relies on the behavior of “==” for the values involved. If you need a more strict test, this can be done in a few lines of code.
Separating tests and source is not hard to do in Python, if you start your source directory at the root of the package structure. This is not hard in principle, but it trips up developers coming from other languages. We review how one popular library does it (which is in line with the pytest docs), … Read more
Python is popular because it is simple to learn and use, with a syntax that makes it as easy to understand as plain English. This ease of use together with highly optimized libraries written in C has made it a favorite in the scientific community, especially for data science and machine learning. Python is not … Read more
A Python package is simply a directory, usually containing Python modules (source files with the .py extension). It frequently will include a special marker file, __init__.py. Packages can easily be created and used locally, and published packages can be installed using pip. Learning to install public packages and create packages locally is not difficult, though … Read more
Yes, it’s true that Python is slower at runtime than some other languages, but it’s fast enough for the tasks for which it is commonly used. Yes, it’s true that Python is slower at runtime than some other languages, but it’s fast enough for the tasks for which it is commonly used. For the most common tasks, the popular libraries use well-optimized C code — but you don’t need to know C to use them. We separate the facts from the myths.
