Python Weekly (Issue 596 April 20 2023)

Python Weekly - Issue 596

April 20, 2023

Python Weekly

Welcome to issue 596 of Python Weekly. Let's get straight to the links this week.

From Our Sponsor 

found by Snyk scans of Python apps. Learn more about how these high risk vulnerabilities might be impacting open source packages you are using today and how to fix them.

News

The European Union's proposed Cyber Resilience Act and Product Liability Act may have unintended consequences for the Python ecosystem. The PSF is concerned that overly broad policies will unintentionally harm the users they are intended to protect. For example, the proposed laws could make it more difficult for open source projects to be updated and maintained, and could lead to increased legal costs for developers. The PSF is calling on the EU to reconsider these proposals and to work with the open source community to develop policies that are both effective and fair.

Articles, Tutorials and Talks

This post discusses the challenges of productionizing large language model (LLM) applications. It covers topics such as prompt engineering, cost and latency, task composability, and promising use cases for LLMs. The post argues that LLMs can be powerful tools for a variety of tasks, but that they require careful engineering to be used effectively in production.

Your code and its execution traces contain enough information to save, query and version computations without extra boilerplate. mandala is a persistent memoization cache on steroids that lets you tap into this information. This post walks you through mandala to give you an idea of its general spirit and the particular ways in which it simplifies computational data management.

Type hints and annotations are not enough when you are using pandas for data analysis in Python. You need validation! Today I’ll show you how to work with Pandera to quickly and easily validate your dataframes.

The post discusses how to build reproducible Python environments using xars, a tool developed by Jane Street. Xars are self-contained archives of Python packages and their dependencies, allowing for easy distribution and reproducibility across different machines.

This article shares the author's experience in building a Python REPL augmented with ChatGPT. He explores how the application is built, and speculates on software engineering patterns and paradigms that might emerge in systems built on Large Language Models (LLMs).

The post describes improvements made to the Python API of GDB, a popular debugger. Specifically, the post discusses improvements to the type and message handling within the API, making it easier for developers to use GDB in their Python projects.

It turns out that the fastest way to create a Pandas DataFrame from a CSV file is to use an entirely different library.

This post is a brief tour of how to set up and use Polars, working through a pragmatic example of the things you might do in a typical data exploration project.

Algebraic data types (ADTs) are a way of representing data that can be in one of a few different, mutually exclusive, forms. In Python, ADTs can be represented using the typing module. ADTs can help to make code more clear and concise, and can also help to prevent errors.

Interesting Projects, Tools and Libraries

State-of-the-art, production ready bots made mega-easy, so you don't have to build them from scratch.

Polyfactory is a simple and powerful mock data generation library, based around type hints and supporting dataclasses, typed-dicts, Pydantic models and more.

Bringing large-language models and chat to web browsers. Everything runs inside the browser with no server support. 

openai_pricing_logger is a Python package that logs OpenAI API costs and timestamps. It's designed to help you keep track of API usage and related costs, providing an easy way to monitor and analyze the expenses.

txtinstruct is a framework for training instruction-tuned models.

Automatically execute code blocks within a Markdown file and update the output in-place.

Automating scientific workflows with AI.

LLM Zoo is a project that provides data, models, and evaluation benchmark for large language models.

A reverse engineered Python API wrapper for Quora's Poe, which provides free access to ChatGPT, GPT-4, and Claude.

A full-body keyboard using gestures to type through computer vision.

Give your python scripts regenerative healing abilities!

Observable Plot in Jupyter notebooks and Quarto documents.

New Releases

This is the second release of pip for the year 2023. You can find the full list of changes in the

.

Upcoming Events and Webinars

There will be following talks

  • Play in your neighbor's backyard for fun! Different projects, same methods

  • IBM Call for Code, Sustainability, and AI (Watson Assistant)

  • Predicting temperatures using linear regression with Python

There will be a talk, Design and Analysis of Experiments with R and Python.

Our Other Newsletters

 - A free weekly newsletter for programmers.

- A free weekly newsletter for entrepreneurs featuring best curated content, must read articles, how to guides, tips and tricks, resources, events and more.