Python News, Tutorials, and a new programming language for AI

Python news and tutorials from last month

Hi friends,

Welcome to another edition of my monthly newsletter. Let’s dive right in:

❓What can you expect?

This newsletter features tutorials, news, cool repos, and educational nuggets related to Python & AI.

⚡️Sponsor

Build a software business faster in pure Python with PySaaS. PySaaS a boilerplate Python codebase that takes care of the fundamental structure and components standard to all SaaS applications such as accepting payments, authentication, an integrated blog, and a nice landing page.

📺📚Tutorials

Here’s a short selection of interesting tutorials from the last weeks that are worth watching:

• Harvard CS50’s Introduction to Programming with Python is now available for free on freeCodeCamp’ YouTube channel.

• Master PDF Chat with LangChain from Sam Witteveen.

• Vector Databases simply explained in 5 minutes. A video I created for the AssemblyAI channel.

• The ultimate introduction to modern GUIs in Python with tkinter. An extensive 18h course from Clear Code.

• DeepLearning.AI launched 3 new courses that are available for free for a limited time: LangChain for LLM Development, How Diffusion Models Work, and Building Systems with the ChatGPT API

• Article: How to split a List into equally sized chunks in Python. Python 3.12 will get a cool new method to do this for you!

🐍 News from the Python Ecosystem

• Chris Lattner announced a new programming language called Mojo🔥, which builds on top of Python and is designed specifically for AI applications. They claim it’s up to 35000 times faster than Python. I made a video about it here.

• PyPI was subpoenaed. They published a blog post about it here for transparency. They also announced that every account that maintains any project or organization on PyPI will be required to enable 2FA on their account by the end of 2023.

• In April, the pip team released Pip version 23.1 which solves dependencies much faster thanks to an improved backtracking algorithm. I recommend upgrading to the newest version with python -m pip install --upgrade pip.

• The open source library LlamaIndex raises $8.5M seed round to build out the data framework for LLMs.

🤝Cool Repos

• lit-parrot: Hackable implementation of state-of-the-art open-source large language models such as StableLM and Falcon.

• mlc-llm: a universal solution that allows any language models to be deployed natively on a diverse set of hardware backends and native applications.

• MiniGPT-4: Enhancing Vision-language Understanding with Advanced Large Language Models

• starcoder: StarCoder is a language model (LM) trained on source code and natural language text

• assemblyai-python-sdk: Finally an SDK from AssemblyAI that makes it super simple to transcribe and analyze audio files.

💡Python Tip

The print function can take additional arguments such as sep, end, and file:

# 1. sep defines the separator between all objects, default is a space

print("How", "are", "you", sep="-")
# How-are-you

# 2. end defines the character in the end, default is a new-line char

print("Hello", end="")
print("World")
# HelloWorld

# 3. file to write the output into a file instead of stdout

f = open("test.txt", "a")
print("This goes into a file", file=f)
f.close()

I hope you enjoyed the read!

Patrick