NumPy 2 is coming: preventing breakage, updating your code

date: 2024-01-16 · Tags: #python, #dev, #news

I read a article on Python Speed about the upcoming NumPy 2.0 release. It shows the new features and changes that are coming in the next major release of NumPy. The article[pythonspeed-for-np2] and offical migration guide[np2migration][np2changelog] is a good start if you are a library developer with NumPy or using NumPy in your projects.

Then I just realized that the NumPy 2.0 is coming. I'm excited to see that.

Check the [numpy/numpy issue #24300]¹ for the latest updates on the development status and announcements.

Footnotes

1. NumPy 2.0 development status & announcements #24300 ↩

SIMD in Pure Python

date: 2024-01-13 · Tags: #python, #simd, #good-reading

1. Python Has infinite precision integers (Python also has a bytearray type)
2. Python has bitwise operation (&, |, ^, ~, <<, >>)
3. Leverage compiler optimizations (CPython compiles your code into a bytecode format)

That's enough to implement simple SIMD operations in pure Python.

Awesome!

• SIMD in Pure Python
• Bitslicing, An Introduction

Representing SHA-256 Hashes As Avatars

date: 2024-01-09 · Tags: #reading, #ideas

Awesome idea to represent SHA-256 hashes as avatars.

Examples:

The above SVG represents the SHA-256 hash of the string "Hello, world!" (in hexadecimal) as an avatar. The hash is:

315f5bdb76d078c43b8ac0064e4a0164612b1fce77c869345bfc94c75894edd3

Super cool :lol

Refs:

• Representing SHA-256 Hashes As Avatars
• Hashvatar Playground