Writings
写作 Writings
共 9 篇。主题涵盖 Python 爬虫、数据分析、Android、Hexo 与工程工具。 9 posts on Python crawlers, data analysis, Android, Hexo, and engineering tools.
拥抱AI时代Embracing the AI Era
专业可以跨界，但底子不能丢。从工具到基础设施，从产业竞争到用、做、想——写下我对拥抱AI时代的判断。You can change fields, but not abandon the foundation. From tools to infrastructure, from industry competition to use–build–think—why embracing the AI era matters.
Tmux的基本使用Getting Started with Tmux
Tmux是一个优秀的终端复用软件，可以在一个终端窗口中运行多个终端会话，并支持后台运行与按需接入。Tmux is a terminal multiplexer for running multiple sessions in one window, with detach/reattach when you need it.
住房月租金预测大数据赛总结Rental Price Prediction Contest Write-up
第一次参加数据分析比赛的复盘：流程、收获，以及改进后在 A 榜冲到第 3 名的模型尝试。A first data contest retrospective: the pipeline, lessons learned, and a model that later reached 3rd on leaderboard A.
成功解决在hexo中无法显示数学公式的问题Fixing Math Rendering in Hexo
Hexo 默认渲染数学公式容易出错；更换 Markdown 引擎并完成配置后恢复正常显示。Default Hexo math rendering often breaks; switching the Markdown engine and config fixed the formulas.
破解含语序问题的点击验证码Cracking Sequence Click Captchas
针对按语序点选汉字的验证码：YOLO 定位、切割、分类识别，再组成点击序列。For Chinese click-captchas in reading order: YOLO detection, cropping, classification, then building the click sequence.
体验活动的生命周期Exploring the Android Activity Lifecycle
通过小实验观察 Activity 四种状态与七个回调，弄清页面切换时系统如何调度。A small experiment on Activity states and the seven callbacks—how Android schedules screens as you navigate.
Hexo博客迁移到一台新电脑Migrating a Hexo Blog to a New Machine
换电脑后用分支管理 Hexo 源文件，把半途而废的博客环境重新接起来——不忘初心。After switching laptops, used a branch to manage Hexo source files and resume the stalled blog setup.
基于Scrapy框架的CrawlSpider类爬取当当全网商品信息Scraping Dangdang with Scrapy CrawlSpider
用 CrawlSpider 抓取当当商品与评论，写入 MySQL，并用中间件应对访问频率限制。Used CrawlSpider to collect Dangdang products and reviews into MySQL, with middleware for rate limits.
基于scrapy框架爬取西刺代理并验证有效性Scraping and Validating Xici Proxies with Scrapy
爬取西刺代理并做有效性验证，把可用代理输出到 JSON，降低被封 IP 的风险。Crawled Xici proxies, validated them, and saved working ones to JSON to reduce IP bans.