MooS ~ 模型即服务?!

原文: PyCoder's Weekly - Issue #566


  • 230301 Zoom.Quiet(大妈) 用时 37 分钟 完成格式转抄.
  • 230301 Zoom.Quiet(大妈) 用时 13 分钟 完成快译

Are you still using loops and lists to process your data in Python? Have you heard of a Python library with optimized data structures and built-in operations that can speed up your data science code? This week on the show, Jodie Burchell, developer advocate for data science at JetBrains, returns to share secrets for harnessing linear algebra and NumPy for your projects.





While multiprocessing allows Python to scale to multiple CPUs, it has some performance overhead compared to threading. This article details why processes have performance issues that threads don’t, ways to work around it, and a sample bad solution.


没事儿, 用 Rust 重写一遍就好...


XML and YAML are two of the most popular text based data formats. This article teaches you how to use third-party Python libraries to convert from one to the other.



None ...


Articles, Tutorials and Talks

Text annotation is the process of reading natural language data and adding additional information to it in a way your program can use it. This info can be used to train models or help process the data. This article describes 6 different tools that can help you annotate your text data.


AI 饲料厂....


Find your favorite package and turn to the readme to get it installed - it seems dead simple just a ‘pip install’ away. Nothing could possibly go wrong. Right? If you’re used to it, it is easy to forget almost all the instructions are skipping a step: using a virtual environment.


历史遗留问题, 一堆堆的可用方案, 每一个都有自己不适合的场景...

从气势来看 PDM 是个狠人: pdm-project/awesome-pdm: A curated list of awesome PDM plugins and resources


In this tutorial, you’ll learn how to iterate over a pandas DataFrame’s rows, but you’ll also understand why looping is against the way of the panda. You’ll understand vectorization, see how to choose vectorized methods, and compare the performance of iteration against pandas.

App Fiddle is to apps what JSFiddle is to JavaScript. Use this instance to learn Flask/SQLAlchemy, running an app in Codespaces. You can browse and explore using VSCode on the web, customize, and debug a complete project, including a database.

This is Duarte’s take on what tools and practices to use for a new Python project. Includes samples for pyproject.toml, details on using pip-tools, and even the occasional Makefile.


和其它社区规定了统一的最佳工程模板相比, Pythonic 世界每个人都可以打造自己最舒服的模板... 虽然舒服, 但是, 对于团队协作就有点...嗯嗯嗯了...


Metaclasses are part of the darker corners of Python and many developers avoid them. This article dives deep into how you can use them to reduce boilerplate code and build APIs.

This step-by-step guide shows you how to build a REST API with Create, Read, Update, and Delete methods using Flask, SQLAlchemy, Postgres, and Docker.


Flask+Pg+Docker ....

全部是重型武器, 讲真一个 Nginx 套个 Pg 本身就可以完成这种服务的发布吧?


Everybody is talking about GPT, this article is actually building one. Learn how to implement a GPT model from scratch in NumPy.


Interesting Projects, Tools and Libraries, Projects & Code


虽然是玩具级别的, 但是, 相比 Docker 一个 .zip 文件是真友好那.




不是, 这是为什么呢? 为了卖的更好?






传统人力自动内容生成器, 不过,作者的 Youtube 宣传片儿制作的真顺溜...

很难相信, 这种可以海量自动生成的短视频内容, 就是很吸粉...



MooS ~ Model-as-a-Service, 小破球的梗儿在这儿接住了...

--> 用户协议 · 魔搭社区

... 由阿里巴巴达摩院,联合 CCF开源发展委员会,共同作为项目发起方;


📆🐍 活动/大会

Events, MeetUp 真的是全球线下活动组织中心


❤️ Happy Pythonic ;-(大妈私人无责任播报)


Happy Pythoning!

Copyright © 2023 PyCoder’s Weekly, All rights reserved.



开始有小伙伴加入承担 pythonisa 周刊的翻译, 从来没提醒过, 可就这么默默坚持下来了...


[皱眉]每周新闻资讯 怎么能错过 
    what f**k 还能这样玩? 还有这东西?

无法同意更多... 很多社区贡献看起来辛苦, 其实受益最多的, 就是主动承担者也.

好文笔,感叹号年度配额: 2/3


(邮件列表地址, 当成正常邮件发送邮件就好, 不用注册, 不用翻越...)


就是四处 是也乎,( ̄▽ ̄) 的那个大妈:

私自嗯哼: ZoomQuiet (订阅号: ZoomQuiet42)
公开课程: 蟒营 (订阅号: Mainium)
历史吐糟: Chaos42 (订阅号 PythoniCamp)

as 创始组织者:
    PyChina (订阅号: PyChinaOrg)
        GDG珠海 (订阅号: GDG-ZhuHai)
        TFUG珠海 (订阅号: ZH_TFUG)



关于 ~ DebugUself with DAMA ;-)
点击注册~> 获得 100$ 体验券: DigitalOcean Referral Badge

订阅 substack 体验古早写作:

关注公众号, 持续获得相关各种嗯哼:


**2021.01.11** 因大妈再次创业暂停定期开设, 转换为预约触发:
  • + 扫描预约入群, 学员每满 42 人即启动新一期训练营 ;-)
  • 101camp22.7
  • + 任何问题, 随时邮件提问可也: