This quickstart guide will get you up and running with Python so you can participate in the Portland Data Science Group. This will be done in three steps:
- Set up a Python development environment
- Choose a resource to learn Python
- Get started with
pandas(Python Data Analysis Library)
This isn't a comprehensive tutorial, just a starting point. Your goal here is to go from your first Hello world! program to what Chris Moffitt calls the plateau of productivity. Many people who start programming for the first time get stuck in the "trough of disillusionment" stage of learning, and struggle to break out of it.
Be prepared to spend a few hours of your own time each week on Step 2, which should be ongoing until you completely finish one of the learning resources. To avoid getting stuck, work through each example by actually typing them out in your Python interpreter — it's important to develop an intuition and muscle memory for programming, so that you can build speed and confidence in your skills as you learn.
The very first thing you'll want to do is download the Anaconda distribution of Python 3, and start up an IDE of your choice via Anaconda Navigator. I recommend Spyder or Jupyter Notebook or JupyterLab for running through exercises.
Note: Anaconda 5.1.0 is missing a dependency for JupyterLab. If you want to use it, open the Anaconda prompt and type conda install nodejs before trying to launch JupyterLab.
There are five main resources recommended by the Python community, all of them free. You only need to choose one and stick with it, but as we are all blessed with different minds, you'll want to try several to see which is more suited to your learning style.
- Automate the Boring Stuff with Python by Al Sweigart
- Dive into Python 3 by Mark Pilgrim
- Think Python by Allen B. Downey
- Python 101 (ebook version here) by Michael Driscoll
- The Official Python Tutorial by the Python Software Foundation
All of them cover the basics of Python, but don't completely overlap with each other in the topics they cover. The main difference is that 1 & 2 get you off the ground faster, giving you the knowledge to start writing practical programs very quickly, while 3-5 are like textbooks and take longer to get through, but are more comprehensive and focus on helping you build a solid foundation in Python. It's completely fine to go with one of the faster resources and fill in the gaps later when you can accommodate it.
I like "Automate the Boring Stuff with Python" the best, which took me about 50 hours (and numerous tries) to get through completely. At some point early on, you'll start to get the itch to write real programs. That's perfectly normal, and I encourage you to start applying what you've learned using a project-based approach — finding problems and solving them with Python is the best way to remember what you've learned so far, and ultimately get better at programming, rather than spending all your time reading and watching videos.
The only resource I recommend avoiding is "Learn Python the Hard Way" by Zed Shaw, mainly because it's outdated.
Once you've done a few chapters of learning Python and start to feel comfortable working in it, you can dive into pandas using Greg Reda's 3-part introduction — it should make sense to anyone who's used Excel or SQL. Once again, you'll want to start up your Python interpreter and type out the examples to get a better understanding of what pandas is doing and how to do it yourself.
Once you feel that you understand how pandas works, try doing some excercises on your own. I recommend one of Chris Moffitt's examples below:
- Excel “Filter and Edit” - Demonstrated in Pandas
- Learn More About Pandas By Building and Using a Weighted Average Function
If you found this introduction too easy or didn't leave you with a strong impression of how to work with pandas, check out Tom Augspurger's 7-part series on writing modern idiomatic pandas code:
As you learn Python, you'll start to understand why styling and whitespace are important. David Goodger's "Code Like a Pythonista" and similarly PEP8 are important companions to have so that you get into the habit writing readable code. While you can afford to skip over some bits of Python as a language to learn at a later time, you can never afford to write crappy code because it is the opposite of saving time.
As SICP wisely stated:
[...] a computer language is not just a way of getting a computer to perform operations but rather [...] it is a novel formal medium for expressing ideas about methodology. Thus, programs must be written for people to read, and only incidentally for machines to execute.
Or as The Zen of Python simply says: Readability counts.
PB Python is a monthly-ish blog by Chris Moffitt (@chris1610) where he talks about how to apply Python in an everyday business setting. This is the best resource I've found for contextualizing how Python can be used as a powerful tool for enhancing or eliminating tasks that are normally done using Excel or Word. It's also a great place to get started with some small pandas projects to understand how to use it for everyday data analysis.
Head into the PB Python Archive and start with the article from 5 October 2014, "Using Sets for Data Analysis". As you move forward, go through the articles in chronological order. Chris does a fantastic job of building upon each one until you have a formidable set of tools to tackle a variety of problems!
Note: For planning purposes, each article can take 1-2 hours to get through, depending on the topic and your skill level.
Regular Expressions 101 by Firas Dib
Regular expressions are a powerful tool, but can be hard to dive into while learning Python. This regex tester and repository is a great resource that can get the regex you need into your Python applications, while also explaining how a specific regex works.
This text is perhaps the most comprehensive resource for beginner Python users, and covers some topics that aren't in any of the Learning Python for the First Time resources, such as interfacing Python with C/C++. This is of particular interest for engineers that work with embedded systems that require C/C++, or anyone that desires the computational power of C/C++ while keeping the ease of development in Python.
Since this book is a behemoth of over 1,600 pages, I recommend it mainly as a companion reference for looking up concepts & examples that aren't explained thoroughly by any of the other resources. For even more exercises, Mark Lutz made his live course materials available for free.
By the time you've completed some programming projects on your own, you'll probably need some resources to reach the next skill level as a Python developer. The books in this section will help correct some of the bad habits you've undoubtedly picked up while learning Python, and fill in any knowledge gaps while also delving into more advanced programming methods. "Fluent Python" is a solid and frequently recommended text that you can't go wrong with, but each of these books has something different to offer depending on what your learning goals are.
Full Stack Python is a landing point for intermediate Python learners looking for a specific focus or application. This open book includes a wealth of resources for getting into web development, data analysis, security, and application deployment, which can be helpful for coming up with projects and learning goals beyond mastering the language.
From the author of "Effective Python" on the differences from "Python Cookbook":
Author of Effective Python here. I think they're very different books! David Beazley is awesome and a wonderful educator. He's always the best speaker at PyCon. I'd say: the Cookbook is a powerful and thorough reference, the Effective books are short and scenario driven.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent