How to sort anything

by - Reuven Lerner

Sorting is one of those things that we take for granted in Python. The built-in "sorted" unction knows how to sort any iterable of objects that are themselves sortable.

But hiding behind that simple description is a great deal of depth. In this talk, I'll go deep into what it means to sort, and how we can sort any collection of Python data. We'll see how you can use custom functions to sort built-in data structures in new and interesting ways. And we'll see how you can design your own custom classes such that they will sort in just the way you want.

After watching this talk, you'll have a better understanding of sorting, built-in data structures, function objects, and how "magic methods" affect the our Python classes. Moreover, you'll be able to write clearer, shorter, and more easily understood code.

Automating your daily tasks!

by - Manas Najmuddeen

A ton of automation can be done by knowing a little bit of python. Some of the everyday tasks you can start with are working on files (read/write), sending emails, reading from your favourite APIs or scraping websites. This talk will give you the "know-how" to start your own journey in Python Automation.

Introducing FastAPI for Python Web Development

by - Tonya Sims

“All the cool kids are using FastAPI for web development.” Imagine hearing this just as you start getting more comfortable using other frameworks, like Django or Flask, to build your web app.

There's a new kid on the block. In this talk, let's walk through how to build a simple web application with FastAPI that transcribes sports audio in real-time.

What is Fast API?

Fast API (aka the new kid) is a modern Python framework that takes all your favorite features from other tools and combines them into one. It's built for speed, rapid development, and enhanced developer experience.

By the end, you'll have a much better understanding of how to use FastAPI in your next project.

This talk is for developers of all levels, including beginners.

Gradual Typing and Django

by - Kracekumar

Python introduced gradual typing with the mypy tool in 3.5. It has received mixed opinions in the community. As the code base grows over time, typing becomes an important tool to maintain code-quality and improve developer productivity.

The talk will introduce the audience to gradual typing in Python and focus on adding type-hints to Django using Django stubs, a third party plugin.

Hybrid Machine Learning and Reasoning Systems

by - Mirantha Jayathilaka

Machine learning techniques such as deep neural networks have gained massive popularity in the area of artificial intelligence in the recent past due to their high accuracy in narrowed-down visual and language tasks. But the latest research findings demand the need to address their limitations such as low explainability and high data dependency, especially when it comes to high impact applications. When the evolution of the study of artificial intelligence is considered, we find that techniques such as deep neural networks fall under an area named sub-symbolic or data-driven machine learning. There exists another realm of techniques that use logic rules and explicit, highly interpretable representations named symbolic or knowledge-driven artificial intelligence. This talk introduces a family of techniques named hybrid machine learning that combines these two areas. The aim of these hybrid approaches is to reap the benefits of symbolic approaches to overcome the limitations of state-of-the-art data-driven machine learning.

Building an end-to-end Data Product using Python

by - Bargava Subramanian

Cricket is a religion for more than 1 Billion people in the world. But as fans, the kind of analytics that’s available and the tools to do further analysis on his/her own has been quite limited so far.

In this talk, the speaker talks about how he and his team created a data product to analyze Indian Premier League (a Twenty20 league based in India). The resulting app, iplnani.com was entirely built just using Python.

He covers how the data was sourced and how a web product was built using streamlit, a simple yet potent library to quickly build web applications. The analysis was done using pandas, the most popular data analysis library. The visualizations were built using altair and folium.

The attendees can learn how to quickly build a web app and/or an internal tool for their own use case, be it as a student or for their industry, just using Python.

Taming Nondeterminism with Dependency Injection

by - Haki Benita

If you ever had a test that one day just started to fail unprovoked, or a test that fails once in a while for no apparent reason, it's possible your code is relying on something that is not deterministic. Dependency injection is a way to tame nondeterminism in code, and take back control over your code! As a bonus, it will also make writing tests much easier! In this talk I'll demonstrate how to implement dependency injection in Python and how it can simplify your tests.

Pydon'ts: Write Elegant Python Code

by - Rodrigo Girão Serrão

In this talk I show what idiomatic Python code looks like and illustrate how the subjective notion of “beautiful”, or “elegant” code, can make your code objectively better, regardless of your Python skill level.
We do that by visiting several vanilla Python features that people sometimes forget to learn, as they rush to importing all the shiny modules and frameworks they really want to use. In order to achieve this, we go through the learning journey of a hypothetical Pythonista that wanted to implement `enumerate` from scratch. We will start with a first version written by them and then refactor it incrementally through the use of Pythonic idioms and patterns.
The talk will build around a series of articles available online (at https://mathspp.com/blog/pydonts), where I explore all the vanilla Python features that make Python one of the most interesting programming languages to learn. There are no prerequisites for this talk, and both advanced users and beginners alike can benefit from it.
After the talk, you will have a better understanding of several concepts such as iterables and (infinite!) lazy generators, and you will know how to use them effectively to get the job done.

Building a clean, maintainable and tested code base

by - Julian Camilleri

An introduction to how to write clean code (type hinting, docstrings, inline comments et cetera) and supporting the clean code with maintainable project structure and unit tests.

Goals - with comparisons/examples:

• What's clean, maintainable code? (type hinting, docstrings, inline comments et cetera)
• What's a clean, navigatable project structure?
• Should the tests package be nested into each package?
• Should my 'tests' package structure mirror the project structure? (for easy navigation of which module/package is being tested)
• (Testing) What should I test and what shouldn't I test?
• (Testing) When should I mock and when shouldn't I?
• (Testing) Writing clean test cases
• (Testing) Using pytest to utilize `parametrize` to test multiple combinations, whilst keeping a clean codebase

Introduction to Django REST Framework

by - Chamindu Munasinghe

Django is a powerful framework that supports clean, rapid development of web applications. If you built your web application in Django and need to provide an API, Django REST framework is your best choice. This session will be an introduction on how to use the Django REST Framework.

Introducing Asynchronous SQLAlchemy

by - Sebastiaan Zeeff

Driven by the immense popularity of asynchronous frameworks, such as FastAPI, asynchronous database support became a hot topic in the Python community. As talking to your database often forms a significant portion of the input and output of your application, it's important to do that asynchronously as well. With the release of version 1.4, SQLAlchemy added support for Asynchronous I/O for both its core and ORM features. This means that you can now use the popular SQL toolkit for Python in your asynchronous applications.

In my talk, I will introduce you to the newly added support for asyncio in SQLAlchemy. I will go over how easy it is to set up the new AsyncEngine, point you to some of the differences with synchronous SQLAlchemy, and show you some of the pitfalls to avoid. In addition, I will also show you how you can run your existing synchronous SQLAlchemy code asynchronously to help you transition to a fully asynchronous application. By the end of my talk, you should know enough to start your own asynchronous SQLAlchemy adventure.

Django with PostgreSQL superpowers

by - Paolo Melchiorre

Every time we’re going to create a new project with Django we make assessments on its requirements to choose the best architecture, of which, the database is usually the core.

Django is a database-agnostic web framework but natively supports only4 Open Source databases: PostgreSQL, SQLite, MariaDB and MySQL. PostgreSQL has the richest feature set of any supported database and some of these features are natively supported directly in Django via its contrib module.

In this talk we’ll see how to use to our advantage the features of PostgreSQL as a database in Django, its exclusive features present in its contrib module and also other superpowers that can be exploited through the use of third-party packages.

From functional to differentiable programming in Python with JAX

by - Yasiru Ratnayake

Differentiable programming is a new software paradigm wherein computer programs can be parametrized and optimised to allow them to learn dynamically from examples. The workhorse behind this is automatic differentiation, whereby gradient-based optimization can be brought to bear for such programs efficiently.

In this talk, I will motivate differentiable programming and illustrate how automatic differentiation works by translating between the mathematical framework and Python. While many mature options exist in Python for automatic differentiation, spurred by active work in deep neural networks, Jax is a package of particular interest bringing together a familiar NumPy-like API with flexible, performance-geared compilation.

Throughout, we will survey how functional programming is a natural fit for this new kind of computation as its ideas appear in our own implementations and in Jax. Coming in with familiarity with NumPy, by the end of the talk you should have a basic grasp of automatic differentiation and of Jax to begin exploring its growing ecosystem.

Word Embeddings for Text Classification

by - Nabanita Roy

Natural language is about texts, alphabets, words, sentences, numbers, special characters, emoticons - in a nutshell, every way we express our thoughts. But machine learning is about only and only numbers. This use-case walk-through focuses on an advanced technique that helps us represent natural language as numbers, called word vectors or embeddings, to apply machine learning and build text classification models.

Data lifecycle with Python

by - Sahan Bulathwela

While the world is transitioning from the industrial revolution to the new data revolution, data science and machine learning practices have matured over the last few years leading to more understood and robust realisations of data science lifecycles. The former part of this talk will introduce the ideas relating to how one can improve data drive in a business and outlines briefly where python can be utilised in the data science lifecycle in a business taking the directors' perspective.

However, building Machine Learning (ML) models entail more uncertainty as the outcomes are not deterministic. This encourages the developer/ data scientist to thread in this path in a much more systematic way, to avoid setbacks. The latter part of the talk will take the engineers' perspective and outlines agile data science practices that will help manage developing ML artefacts. This part will also argue and convince the audience how Python presents one of the most stable and widest ecosystems for data science. A variety of Pythonic tools that can future proof your data science career available to carry out critical processes such as processing data, testing ML models, versioning ML models and monitoring ML models will be presented to the audience with example use cases.