<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Autograd on Text Matrix</title><link>https://txtmix.com/tags/autograd/</link><description>Recent content in Autograd on Text Matrix</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Tue, 21 Jul 2026 20:06:14 +0800</lastBuildDate><atom:link href="https://txtmix.com/tags/autograd/index.xml" rel="self" type="application/rss+xml"/><item><title>SmolML：纯 Python 实现的机器学习库，从自动微分到完整模型</title><link>https://txtmix.com/posts/tech/smolml-machine-learning-library-from-scratch/</link><pubDate>Sun, 17 May 2026 20:02:25 +0800</pubDate><guid>https://txtmix.com/posts/tech/smolml-machine-learning-library-from-scratch/</guid><description>&lt;h2 id="这篇文章在讲什么">这篇文章在讲什么&lt;/h2>
&lt;p>SmolML 是一个只有约 2000 行纯 Python 代码的机器学习库，GitHub Stars 443，MIT 许可证。它不依赖 NumPy、SciPy 或任何外部框架——只用 &lt;code>math&lt;/code>、&lt;code>random&lt;/code>、&lt;code>collections&lt;/code> 这三个标准库模块。&lt;/p></description></item></channel></rss>