<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>mxt.log</title><description>学习记录、算法竞赛与 AI 竞赛实践笔记。</description><link>https://mxt1022.github.io/</link><item><title>CS61A 学习笔记：抽象、函数与递归</title><link>https://mxt1022.github.io/notes/cs/cs61a/cs61a%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0/</link><guid isPermaLink="true">https://mxt1022.github.io/notes/cs/cs61a/cs61a%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0/</guid><description>从高阶函数、递归与数据抽象出发，重新理解程序是如何组合起来的。</description><pubDate>Sun, 10 Mar 2024 00:00:00 GMT</pubDate><category>CS61A</category><category>Python</category><category>函数式编程</category></item><item><title>深度学习学习路线：从张量到完整项目</title><link>https://mxt1022.github.io/notes/ai/dl/road/</link><guid isPermaLink="true">https://mxt1022.github.io/notes/ai/dl/road/</guid><description>一条以动手实践为主线的深度学习路线，覆盖基础、模型、训练与项目复盘。</description><pubDate>Tue, 20 Feb 2024 00:00:00 GMT</pubDate><category>深度学习</category><category>PyTorch</category><category>学习路线</category></item><item><title>机器学习：从任务定义到模型评估</title><link>https://mxt1022.github.io/notes/ai/ml/ml/</link><guid isPermaLink="true">https://mxt1022.github.io/notes/ai/ml/ml/</guid><description>用一条完整工作流串起监督学习、无监督学习、特征工程与评估指标。</description><pubDate>Mon, 15 Jan 2024 00:00:00 GMT</pubDate><category>机器学习</category><category>特征工程</category><category>模型评估</category></item><item><title>天池量化赛道：拆解一个 Baseline 的工程结构</title><link>https://mxt1022.github.io/contests/tianchi-baseline/</link><guid isPermaLink="true">https://mxt1022.github.io/contests/tianchi-baseline/</guid><description>从数据读取、交叉验证、CatBoost 到并行实验，理解竞赛 Baseline 各模块的职责。</description><pubDate>Sat, 02 Sep 2023 00:00:00 GMT</pubDate><category>天池</category><category>CatBoost</category><category>交叉验证</category></item><item><title>Kaggle 猫狗识别：从 CNN Baseline 到误差分析</title><link>https://mxt1022.github.io/contests/kaggle-cats-dogs/</link><guid isPermaLink="true">https://mxt1022.github.io/contests/kaggle-cats-dogs/</guid><description>在 Kaggle Notebook 中完成图像分类训练，记录数据增强、过拟合与评估过程。</description><pubDate>Sun, 06 Aug 2023 00:00:00 GMT</pubDate><category>Kaggle</category><category>CNN</category><category>TensorFlow</category></item><item><title>Kaggle Titanic：第一次完整走过表格竞赛流程</title><link>https://mxt1022.github.io/contests/kaggle-titanic/</link><guid isPermaLink="true">https://mxt1022.github.io/contests/kaggle-titanic/</guid><description>从缺失值、可视化与特征构造开始，建立一个可验证、可提交的分类 Baseline。</description><pubDate>Sun, 09 Jul 2023 00:00:00 GMT</pubDate><category>Kaggle</category><category>特征工程</category><category>分类</category></item><item><title>算法训练记录：从每日一题到可复用题型</title><link>https://mxt1022.github.io/contests/algorithm-training/</link><guid isPermaLink="true">https://mxt1022.github.io/contests/algorithm-training/</guid><description>用持续训练、赛后补题和模板整理，把零散做题转化为稳定的问题识别能力。</description><pubDate>Tue, 20 Jun 2023 00:00:00 GMT</pubDate><category>算法竞赛</category><category>LeetCode</category><category>Codeforces</category></item><item><title>数组基础：访问、修改与常见题型</title><link>https://mxt1022.github.io/notes/algorithm/array-basics/</link><guid isPermaLink="true">https://mxt1022.github.io/notes/algorithm/array-basics/</guid><description>从数组操作的复杂度出发，整理中心下标、轮转数组与旋转矩阵的解题模式。</description><pubDate>Sun, 18 Jun 2023 00:00:00 GMT</pubDate><category>数组</category><category>LeetCode</category><category>复杂度</category></item></channel></rss>