This talk will cover how to host a Wagtail/Django backend running on Digital Ocean with Dokku. And a Next.js frontend running on Vercel. This combination leads to an ultra-cheap solution for a scaleable and fault-tolerant solution for personal projects or startups.
The goal of this talk is to give you a roadmap on a journey to writing stronger code with software testing. How do you actually know if your code really works? How do you know that you didn't break something "over there" when you changed something "over here"? In this talk, we'll demonstrate common problems and solutions with respect to verifying code correctness and improving maintainability.
Maybe you've already started trying to learn testing and some things are still unclear:
- "What is the point of a mock?",
- "What is the difference between patching with pytest vs unittest or using a with-block vs a decorator?"
- "I get stuck writing a test as soon as I go to a nontrivial example."
Make neuroimaging results easier to digest for patients and equip technicians with an enhanced toolkit featuring improved visualization and statistical analysis capabilities through MNE, an open-source Python Library.
Help! I want to run a simple task on a schedule how do I do that? Wait, remember when I said "a simple task", well it just got more complicated. And why isn't the data ready when my job runs?! Oh no, yesterdays job failed...what do I do?
These are all common problems lots of data analysts/engineers and scripters encounter. How do you solve them in a sustainable way? This intro talk is helpful for all who think about job scheduling.
This talk is about an open-source educational project based on Python: blupants.com. My goal is to present it to the audience and hopefully find more people passionate about the topic like me who would be willing to help as open-source contributors. Attendees will learn how to use the REST API to control robots and how to extend it to new platforms and robots.
With the rise in AI, there is more focus on Python dependency management and SCA scanning. Python's dependency management system makes it easy for developers to leave dependencies out of the manifest. This means that almost every SCA tool that relies on a manifest will be wrong. We show how leveraging program analysis techniques one can avoid the pitfalls of these so-called phantom dependencies.
What's the difference between "sudo apt install python3.12" and "wget ... && ./configure && make && make test && sudo make install"?
Do I need to use a virtual environment in my docker container?
I installed python, but it didn't come with pip.
Let's take a look at python on Debian. It runs both your OS and your web app, and it's weirder than you think.
Presenting on an open source project I started and have been working on since three years ago. It is a library named PyPDFForm which has a variety of utilities making processing PDF forms easier with Python. I'll discuss what sparked this idea, give a little coding session to demo some of the library's functionalities, and talk about the future of the library.
Here is a brief layout of what I'm currently planning on for the speak.
* First 15 minutes: Intro. Some background, both personal and about the library. What sparked this idea and just general motivations.
* Next 10 minutes: A coding session which demos some basic but essence functionalities of the library.
* Final 5 minutes: Talk about the future of the library. What are some restrictions it has now. What are some improvements that can be done. And QA.
Some PyPDFForm links:
Ansible is a fantastic automation tool written in Python but sometimes playbook logic gets very complicated to do seemingly simple tasks. In our talk we will show how writing a custom Ansible module is not as hard as you might think and how leveraging Python for more complicated logic lets you write easier to navigate playbooks.
Additionally we will review:
- A light Ansible overview
- Explanations of Ansible modules and plugins
- How testing fits in
In this session we will perform a full exploration of the Snowpark ML Toolkit using dbt Python models on the dbt Cloud without any package installation. You will learn how Snowflake improves on familiar ML libraries like Scikit-Learn and XGBoost to make them scale within its scalable runtime.
During this session you will:
- Start with 50,000 rows and scale to 50 million rows on various data transformations with the toolkit
- Train and predict a dbt Python model with XGBoost
- Switch between SQL and Python transformations in a pipeline where dbt takes care of the boilerplate code when pushing to Snowflake