机器学习`侠`练成记录 Becoming a Machine Learning Practitioner

It’s an opinionated list of core skills that I found useful in the daily work of a machine learning practitioner, in the tech industry. I am not a researcher and am not interested in becoming one, so the list does not go in depth into any active research domain.

需要强调的是，我不是做ML科研的，也不想做科研，所以本文就是于业界做ML的大家交流讨论，并不打算深入任何科研领域，请做科研的大牛轻拍。

I’ll be maintaining this list just as I maintained the learning path for data science in the past here https://www.1point3acres.com/bbs/forum.php?mod=viewthread&tid=76429&extra=&page=1.

It’s not meant to be exhaustive, and we probably don’t need to know them all.

另外这篇内容也并不会多完备。机器学习领域大牛很多，领域很广，应用更是广阔到难以尽数，所以不求尽善尽美，只是抛砖引玉，求交流学习。

Suggestions/discussions welcome.

This is a practitioner’s approach. Researchers: This post is not for you. 不是给ML方向做研究的人；牛人路过就好，不喜勿进。

Introduction to Python or Introduction to Java

Data structure ( 地里关于Berkeley 61B的板块 or Coursera specialization )

Design and analysis of algorithms（ Coursera Part 1, Coursera Part 2）

Database (Stanford archived DB course or Using Database with Python)

Discrete math (Coursera Discrete Math specialization)

Operating System (Book: Modern Operating Systems)

为什么与上一节计算机基础分开来说：因为老是遇到同学说我会numpy啊为什么你说我基础不行。。。。实际上是可以在不太懂基础的情况下楞刷题，或者写些基本能用的代码的。但是稍微深一点的地方会感觉基础知识的缺乏会让人难以在ML道路上上升到一定高度。

Must have :Python, Java

必须一门Python（因为ML好多framework就是python)，和一门compiled language.

仅仅只会python，作为scientist是够了，作为engineer就会有显著差距。

Good to have: C++ 不一定是必须的。但是如果做的工作对速度有比较强的要求，那还是需要会C++的

Will likely encounter at work: scala/go/js 这些看每家tech stack不一样，会有不同涉及。遇到了再学即可。

Will likely encounter in academic settings: Matlab/Octave, R, Julia 这些是学校里面用的多，我至今（2018年底）没看见公司里面用julia的。

Linear algebra 线性代数，必须的

Calculus 这个大家应该都上过

Optimization 优化：必须的

Statistics 基础统计（不是概率论 which is also good to have，这里是特指统计）

(real analysis and functional analysis might be useful, but is not required)

有空的话学实分析和泛函也可以，但是不是必须的

- 这门课可能没有人不知道了 Introductiont to Machine Learning by Andrew Ng

- 这本书是ESL的简单版，作为直觉培养和思路练成，仍然是不错的，但是那里面的编程就很轻很轻了，真的只够本科生用 Very light but still a good book: an introduction to statistical learning

- 前面几门主要还是supervised learning，下面这门稍微宽广一点，并不完全是ML，但是也是因此感觉对知识面和落地有帮助

Data mining & other topics: Mining Massive Data Sets http://web.stanford.edu/class/cs246/

- 这门课可惜没有录像，关于实战的部分讲的还是不错的，而且是其他课程都没有涉及到，但是工作里面的确需要的部分

Cornel course (slides only) adv ml http://www.cs.cornell.edu/courses/cs6780/2010fa/lectures.html

- Book:

经典ESL 不必多说，统计角度 Elements of Statistical Learning

经典不必多说，CS角度 Pattern Recognition and Machine Learning

Some intro here:

UIUC course slides https://github.com/SSQ/Coursera-UIUC-Text-Retrieval-and-Search-Engines-Lecture-Slides

Book: Recommender Systems: The Textbook by Charu Aggarwal

Andrej版的CS231n堪称经典 stanford CS231n Convolutional Neural Networks for Visual Recognition

others to be added

(book and course TBD)

- Lots more stuff in DL here

经典课本（但是我觉得读起来还是蛮晦涩。。。不知道我是不是一个人）Deep learning book by Ian Goodfellow and Yoshua Bengio

非深度学习，最常用的肯定就是 General ML framework: sci-kit-learn

深度学习的目前很多了 DL: Tensorflow(Keras), Caffe, Pytorch(Caffe2)

TF的看狗家自己的内容，或者Ng那个课；Pytorch的看fast.ai

Up & coming 或者已经下去了的: theano, MXnet, dl4j

做小数据ML（笔记本上跑跑regression or classification，产生个报告给别人看）严格来说算不上ML，其实主要只能算是modeling（统计建模）

high dimensional data 是另外一个故事，这里先按下不提。

只说业界，面试会考系统设计的地方，需要用到的机器学习系统：

This blog post: thorough intro to distributed systems

And this

Grokking system design interview (for brushing up fundamentals and case studies)

DDIA 堪称经典 Book Designing Data Intensive Applications : for an in-depth look, refer back to fundamental knowledge in OS

- 这门课还是可惜没有录像只有slides

Adv ML systems (Cornell, slides only)

- Book: 这本书我也只有翻过，还不知道到底多好

- NIPS2018: to find some talks http://learningsys.org/nips18/schedule.html

- Prod ML, paper 1 (tech debt) https://ai.google/research/pubs/pub43146,

- paper 2 (test score), https://ai.google/research/pubs/pub46555

- tbd

TFX https://www.tensorflow.org/tfx/ & KDD talk https://www.kdd.org/kdd2017/papers/view/tfx-a-tensorflow-based-production-scale-machine-learning-platform

- Facebook:

FB Learner flow https://code.fb.com/ml-applications/introducing-fblearner-flow-facebook-s-ai-backbone/

Michelangelo https://eng.uber.com/scaling-michelangelo/

深度学习 Horovod https://eng.uber.com/horovod/

Pro ML https://twimlai.com/twiml-talk-200-productive-machine-learning-at-linkedin-with-bee-chung-chen/

Bighead https://databricks.com/session/bighead-airbnbs-end-to-end-machine-learning-platform

另外这个podcast很不错，建议不要错过

Podcast TwimlAI is featuring a lot of these systems lately, a fantastic listen

https://twimlai.com/shows/

其他还有 Amazon, Netflix etc, ...