发现

The dominant sequence transduction models are based on complex recurrent or convolutional neural networks. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms.

A collection of sayings and ideas attributed to Confucius and his contemporaries, compiled by his followers over the two centuries following his death.

Neural net training is a leaky abstraction. It is allegedly easy to get started, but the fast and furious approach does not work and only leads to suffering.

If you collected lists of techniques for doing great work in a lot of different fields, what would the intersection look like? I decided to find out by making it.

curl 是最常用的命令行工具之一，几乎所有系统都预装。它的功能极其强大，但参数多达上百个，很少有人能完全掌握。本文梳理它最常用的那一小部分。