Contents
Occasionally I am asked for advice on how to learn Python, or to get better at Python programming. I've always felt that I'm not necessarily best placed to answer this question, as I learnt Python a long time ago [1], and there are many resources available now that were not even thought of then.
Part of the October 2019 CamPUG meeting was given over to a discussion of Python learning resources. This page is meant to fold together my original notes and the comments and suggestions from that meeting.
I'd love comments / contributions - either via pull requests on this document on github, by email to me directly (again, via github) or by message via the CamPUG meetup page. Don't worry too much about getting the text formatting right - I can always fix that.
-- Tibs
Note: Where this page says "I" it means "Tibs", and where it says "we" or lapses into the passive voice, it may mean information from CamPUG members.
| [1] | in a city far, far away (well, in the early 1990s and in Glasgow, but you get the point). There weren't any Python books or other learning resources back then, apart from the actual Python documentation itself. |
Whatever you do, make sure you're learning Python 3. Any course that teaches Python 2 is much too out-of-date.
Many courses out there seem to be teaching with Python 3.6 - that's OK. Later versions of Python add more nice things, but what you learn with Python 3.6 should continue to be useful.
Do be aware that lots of computers (I'm looking at Mac OS in particular) come
with the python command set up to run Python 2. So you may need to type
python3 to get the right version of Python. Any course you're using should
explain that.
This guide mainly suggests free resources, partly because not everyone has money to spare for learning Python.
Text that is:
"indented and in quotes, like this"
is repeating what someone else has said about their resource.
Different people learn in different ways, so it's worth having a look at several of the following.
Trinket - an online education framework - code in the browser
Trinket lets you edit and run Python code in the browser. There are currently three Python related tutorials on their front page:
- From Blocks to Code, which lets you program by assembling blocks into programs. This also teaches some basics of Python - in particular, after creating the block programs, you can ask it to show you the equivalent Python code.
- Python with Turtles, which uses the Turtle library to draw shapes.
- Python for Everybody, which is a book to teach the basics of Python3, using trinket dialogues for the examples.
Trinket also provide Hour of Python, which includes the turtles and blocks sessions, but also a sequence of challenges and specific tutorials. This website also has at least some resources in Spanish, Chinese and Korean.
scrimba - a live programming environment you use while watching videos. The front page says
"Scrimba lets you play around with the instructor's code directly inside the screencast. As a result, you'll learn faster and have more fun."
If you search for Python then you find quite a few courses, many of which look suitable for beginners. There's a mix of free and paid courses. Their Learn Python for free course is, of course, free.
I recommend the Django Girls tutorial, which starts by teaching what a command line is, and works through to creating a web site using Django (and boys are allowed to use it, too - it says so in the FAQ). They are also creating videos on the "Coding is for Girls" youtube channel, and the tutorial pages link to the appropriate video, if it exists.
You may also find that there is a Django Girls workshop locally, so that you can learn in company with others - see https://djangogirls.org/ for more information.
There are a series of resources at Invent with Python, including free online books and links to youtube videos and paid courses. If you scroll down the page, it gives a brief introduction to each book, and see what looks interesting. Good places to start are either "Invent with Python" iteself, or "Automate the Boring Stuff with Python". There are also books for Lua and Scratch.
realpython.com (Real Python Tutorials) provides regular articles and videos, aimed at all levels of Python skill, from beginner to advanced. You can subscribe to emails, and new posts also appear on Planet Python. The page at https://realpython.com/start-here/ has suggestions about where to start depending on you knowledge and needs, including the Python Basics Book:
"Python Basics Is for Complete Beginners to Python and Coding: No Prior Programming Experience Necessary."
If you've got a Raspberry Pi, then there are lots of resources for using Python on it. One place to start is Raspberry Pi Projects, or just go to https://www.raspberrypi.org/ and start exploring.
Pygame Zero teaches you to write sprite games, which happen to use Python. It does assume you know some basics.
If you're the sort of person who takes comfort in equations and graph drawing, then you may find the Part IA Computing (Michaelmas) course helpful. This is a Jupyter notebook providing a self-study introduction to Python for Part 1A Engineering students at Cambridge University. You need to a Microsoft Azure account to use the notebooks - this is explained in the course notes on the page.
(Note that that link might be moving by October 2020, as Microsoft appear to be making some changes. The source code for that site can be found at https://github.com/CambridgeEngineering/PartIA-Computing-Michaelmas, and if they do move the notebook site, the github page will probably give a link to the new location.)
Finally, don't forget that you can try asking for help on the CamPUG slack chat. You can join at this link (and if that doesn't work, there may be a more up-to-date link in the "What we're about" section at https://www.meetup.com/CamPUG/). Questions about Python use are probably best asked in the #python channel on slack.
If you already have some programming skills, then it can actually be a little bit harder to know what to recommend, because how much one knows can vary so much, and also because people learn in such different ways.
First off, it can definitely be worth looking at the resources in the previous section, Where to start if you don't have much programming experience.
In particular:
- the Django Girls tutorial can be useful for teaching some basic Django (you can skim the early parts that you already know)
- some of the books at Invent with Python may still be interesting
- realpython.com (Real Python Tutorials) really is a good resource for all skill levels - Tibs has definitely found some very useful articles from there.
If you enjoy learning by writing games, then PyGame provides a framework and a lot of resources. There is also Pygame Zero, which provides a simpler way of writing games that is easier to get started with.
I used to recommend Harry Percival's "Test-Driven Development with Python", which also available free at Obey the Testing Goat!. This teaches you how to build a Django website (a more complex one than that in the Django Girls tutorial) by way of teaching Test Driven Development. As such, it does assume some basic Python knowledge, but it also insists that you copy out each example (no cut-and-paste), which should help reinforce knowledge.
The book requires you to use Python 3.6, which is not a big problem, and Django 1.11. Unfortunatley Django 1.11 is now getting a bit old, and you definitely shouldn't use it for Real Work. It also means that getting all of the related software in the correct versions to match may be getting more difficult, and having to sort out installation problems when trying to learn is not so much fun.
On the other hand, it's still a good introduction to the concepts of Test Driven Development.
Finally, as in the previous section, you can always try asking questions on the CamPUG slack chat (join at this link).
When I was learning Ruby, I found the Ruby Koans to be very useful. Their goal is to teach the Ruby language, syntax, structure, and some common functions and libraries, as well as some of the Ruby culture, and basic testing. Other programming languages have adopted this approach (for instance, see the list at Learn a new programming language today with koans), and Python is no exception.
One Python example is by Greg Malcolm (but also see the next paragraph), at https://github.com/gregmalcolm/python_koans. This is partly a port of the Ruby koans (where approriate) and partly new material.
The koans at https://github.com/arachnegl/python-koans are intended to be done before those by Greg Malcolm, as the author says they are meant to be more useful to people of a non-programming background.
The koans aren't always easy. Luckily, if you get stuck, or don't understand the answer, there are normally solutions to be found (via google or duck duck go or your other favourite search engine). And there are other Python koans as well - these were just the first ones I found.
You can use any text editor to work with Python, but it's sensible to use something that at least understands how to indent Python code, and will preferably help point out mistakes as you make them.
If you're already using a text editor that does this, then it's perfectly reasonable to carry on with the same editor for Python.
If you're new to programming, then I recommend the Mu editor. Mu is a Python code editor for beginner programmers based on extensive feedback given by teachers and learners. As such, its core beliefs are:
- Less is More: Mu has only the most essential features, so users are not intimidated by a baffling interface.
- Tread the Path of Least Resistance: Whatever the task, there is always only one obvious way to do it with Mu.
- Keep it Simple: It's quick and easy to learn Mu ~ complexity impedes a novice programmer's first steps.
- Have fun! Learning should inspire fun ~ Mu helps learners quickly create and test working code.
It also makes it very easy to send Python programs to things like the BBC micro:bit and Adafruit boards that run CircuitPython
There are tutorials for how to use Mu at https://codewith.mu/en/tutorials/.
I recommend Mu for pair programming at CamPUG, as it saves discussion of what editor to use, and concentrates the mind on programming rather than clever editor tricks.
If you're wanting something more sophisticated than Mu, then the two best known Python IDEs (Integrated Development Environment, an editing environment specifically tailored to programming), and the ones I tend to recommend, are:
VS Code (Visual Studio Code) is a free IDE from Microsoft which understands Python and lots of other programming languages. It is also meant to be numpy aware.
There is a blog about Microsoft's Python support at https://blogs.msdn.microsoft.com/pythonengineering/
PyCharm from JetBrains.
The community (free) edition is very good, but if you want to do serious Django work then the extra Django support in the paid version is apparently worth it.
The JetBrains blog at https://blog.jetbrains.com/ covers PyCharm and their other IDEs.
At the October 2019 CamPUG meeting on learning resources, mention was also made of Spyder, which is distributed as part of the Anaconda Python distribution. This is an IDE aimed at scientific users.
There is also Atom, which is a free editor developed by GitHub. This is a more general text editor that has optional support for many programming languages, including Python.
You don't necessarily have to write Python programs in an editor on your own computer. There are also online editing environments.
We already mentioned Trinket and scrimba in Where to start if you don't have much programming experience.
repl.it gives you an online IDE, including collaborative code editing, for a variety of programming languages, including Python.
Glitch is a collaborative programming environment in the browser that makes it easy to write web apps in a variety of languages, including Python (https://glitch.com/@python)
Iodide lets you create (scientific) notebooks using Python, markdown, Javascript and CSS, entering text in one pane in the browser, and seeing the result alongside. See A Brief Tour through Pyodide for more information on using it with Python. The project says it is still in alpha, so things may change and break.
(There's obviously some overlap with the capabilities of sites that provide live Jupyter notebooks, although the style seems rather different.)
Anvil lets you write full stack web apps just using Python. That is, both the front end and the back end are both written using Python. There's a drag-and-drop interface for designing the front end. The Anvil team come and give workshops at CamPUG periodically, and can be found at many Python conferences. There are tutorials and documentation at https://anvil.works/learn.
Quoting the Moving on from Mu page, "Jupyter Notebooks are an amazing way to create an interactive narrative with code, multi-media and traditional prose".
Jupyter notebooks can be used as a way of mixing notes (in markdown) and Python code, with the results of running that code. This idea is very much based on lab notebooks, where text, calculations and graphs would all be written down. Only here, the calcualtions and graphs (and other things) can be generated live from the results of running Python code.
(Also, Python is not the only programming language supported, so this is a useful technology to use whatever programming language you may be using.)
There is at least one gallery of interesting Jupyter notebooks out there.
Try Jupyter lets you try out some tutorial Jupyter notebooks in your browser. There are tutorials for several programming languages, including Python.
Microsoft Azure Notebooks provide online Jupyter notebooks, and host various tutorials presented as Jupyter notebooks (including the Cambridge Unversity Part IA Computing (Michaelmas) Python course mentioned elsewhere). The Azure Notebooks homepage has links to various featured projects and tutorials, and there is an informal introduction to Python3.
CoCalc provides an "online computing environment" that supports a variety of things, including Python and Jupyter notebooks. They appear to have an unlimited free trial, which should be enough to experiment.
There are other places on the web where you can store and interact with Jupyter notebooks - this was just a selection.
Many people need a purpose to learn something like Python. For some people this need for a purpose can be satisfied by programming challenges. For instance:
-
From the About page: "Python Challenge is a game in which each level can be solved by a bit of (Python) programming. ... All levels can be solved by straightforward and very short scripts. ... (well, except of level 32)."
This expects you to solve problems by writing snippets of Python code. Sometimes needs you to think quite hard (to figure out what it wants you to do). You're also expected to use some of the standard library, and external libraries like the imaging library pillow (yes, that's a hint for one of the questions).
-
"Project Euler exists to encourage, challenge, and develop the skills and enjoyment of anyone with an interest in the fascinating world of mathematics".
This isn't Python specific. It is a series of mathematics based problems, at many different skill levels, that you can use Python to solve.
-
"Advent of Code is an Advent calendar of small programming puzzles for a variety of skill sets and skill levels that can be solved in any programming language you like. People use them as a speed contest, interview prep, company training, university coursework, practice problems, or to challenge each other."
These come out every December. Again, they're not Python specific, but Python works well. Every day from 1st to 25th December, you get a new puzzle that builds on the previous one, and the challenge is to sumbit a correct solution. The puzzles for previous years (back to 2016) are still there as well (click on
[Events]at the top of the Advent of Code home page), so there's a lot to play with here. The yearly PyWeek challenge:
- Invites entrants to write a game in one week from scratch either as an individual or in a team,
- Is intended to be challenging and fun,
- Will hopefully increase the public body of game tools, code and expertise,
- Will let a lot of people actually finish a game, and
- May inspire new projects (with ready made teams!)
Also see the section on Koans.
And if you think you have a decent understanding of Python, take a look at What the f*ck Python? which is a page dedicated to "Exploring and understanding Python through surprising snippets."
Tibs says:
For interest’s sake, I follow https://www.reddit.com/r/Python/, which sometimes has interesting things on it. There’s also https://www.reddit.com/r/learnpython which is specifically aimed at beginner questions - you’ll have to judge whether the questions and answers there are also useful. I assume it’s worth a look, though.
Planet Python is an aggregator of Python related articles - there are often introductions to interesting things referenced here, so it’s probably worth keeping an eye on (I have it as an RSS feed, altthough these days the reddit channel has more that’s directly of interest to me - but your mileage may well vary).
https://stackoverflow.com/ can, as ever, be useful, but be sure that the answers you’re looking at are for Python 3, and even Python 3.5 or later, or they’re not so likely to be of use.
- As mentioned in Where to start if you don't have much programming experience, realpython.com has useful articles and videos aimed at all levels of knowledge and is worth checking out.
- Similarly, the resources at Invent with Python are worth looking at.
- Python Idioms (2014) is a set of slides by Safe, one of the original CamPUG founders, which gives ten useful Python idioms (commonly used and understood ways of doing things)
- Effective Python (not free) is a book that assumes some knowledge of Python, but tries to suggest the way an experienced Python programmer would do things (often termed the "Pythonic" way of doing things).
- Once you're comfortable with Python, you may be interested in deeper dives into how things work. I think "Fluent Python: Clear, Concise, and Effective Programming" by Luciano Ramalho (not free: Fluent Python on amazon.co.uk) is an excellent book for this purpose.
- And as mentioned before, What the f*ck Python? challenges the idea that one has an understanding of how Python actually works.
Don't forget the official Python 3 documentation [2].
The Python tutorial is worth re-reading periodically, if only to see what has become a part of the "basic" language. As it says, it
"does not attempt to be comprehensive and cover every single feature, or even every commonly used feature. Instead, it introduces many of Python’s most noteworthy features, and will give you a good idea of the language’s flavor and style."
It's probably not ideal as a first place to learn Python from.
Tibs would recommend looking at the first few sections of the library reference, probably Introduction through Built-in Exceptions, and then the chapters on string, datetime, collections, pathlib, os.path and os (yes, those last three are different - blame history), and then it very much depends on what you want to do.
| [2] | this will give you the documentation for the latest version. If that's not what you want, choose the version you do want with the selector at the top of the page. The text itself is normally fairly good at saying which version of Python introduced a new-ish feature. |
If you need to use logging, regular expressions, unicode, argparse (command line parsing) and a few other things, the HOWTO documents at https://docs.python.org/3/howto/index.html are actually a better place to start.
The Python Tutor - Visualize Python, Java, C, C++, JavaScript, TypeScript, and Ruby code execution is an interesting page that allows you to type in Python (or other) code and then see how it executes. I've had someone say it can be "especially great for learning recursion".
One way to learn more is to attend some of the various meetings.
- CamPUG itself meets once a month, normally on the first Tuesday of the month. We continue to meet virtually
- Cambridge PyData also meets once a month, normally on the last Wednesday of the month. It continues to meet virtually, and (at least sometimes) combines its meetings with other PyData meetups.
- Before the pandemic, the Raspberry Pi Foundation hosted Raspberry Jams, sometimes in Cambridge. Hopefully they will resume sometime in the future.
- Also sadly postponed are the International Raspberry Py Robotics Competitions, or Pi Wars, which happen in Cambridge over a weekend.
- London has its own PyData London Meetup and also the London Python Coding Dojo - see their websites for details.
One way to keep up with Python events is to subscribe to the UK Python Users mailing list, at https://mail.python.org/mailman/listinfo/python-uk
- In a more normal year, we would also expect to see the UK Python conference. As it says at https://2020.pyconuk.org/, that's sadly not happening in 2020.
- Similarly, there was meant to be a Cambridge PyData 2020 conference, which has also had to be postponed.
- PyLondinium is the London Python conference, and it also postponed the 2020 event.
- PyCon is the biggest Python conference, normally in the USA. PyCon 2020 did go ahead, but as a virtual conference, in April 2020.
See some stuff about going to conferences at Notes about conferences
There are lots of videos out there. In particular, the PyCon and PyCon UK conferences put a lot of their talks up on YouTube. Have a look at
- The PyCon UK channel (remember, there is/was no PyCon UK in 2020)
- The PyLondinium channel (also, didn't happen in 2020, but there are videos from 2019 and 2018)
- The PyCon 2020 channel and the PyCon 2019 channel (and you can surely work out how to find previous years)
Both conferences have talks at all technical levels, as well as talks about things that aren't directly to do with programming. Lightning talk [3] sessions can be especially worth watching.
| [3] | Short talks, of less than 5 minutes, on any topic at all. |
People sometimes ask about how to improve the performance of Python programs. This depends a bit on what sort of program it is, but here a are a few hints:
- High Performance Python by Ian Ozsvald and Micha Gorelick (also available at amazon) is probably the best introduction to this topic.
- Unsurprisingly, the Cambridge PyData meetup is a good place to ask questions about making Python run faster, or finding out better ways to approach the sort of problem that makes people ask this question.
- Software Carpentry and Data Carpentry are low cost courses for academics
- One of the well known approaches is to use Cython, which gets you closer to the performance of C whilst still writing code in something close to Python.
These notes and any related files (i.e., anything in this repository) are released under a Creative Commons Attribution-ShareAlike 4.0 International License.
