Skip to content

Latest commit

 

History

History
52 lines (34 loc) · 2.02 KB

topic_002.md

File metadata and controls

52 lines (34 loc) · 2.02 KB
Raw

Overview: Why big picture is relevant

  • Data Science == Programing, Statistics, Math, Domain Specialist, Business specialist
  • Programing == Data engineer, code optimizer, Object Oriented coder, Scientific Programing
  • Statistics == Bayesian, Frequentist, Statistical Machine Learning
  • Math == Numerical Linear Algebra, Wavelets, Signal processing

Breakout One

This will be the most open ended of our work this semester and it will give me a chance to get to know you and to know each other.

It’s fairly standard to thinking of the following specialists positions in data science. (Programer, Statistician, Domain expert/Business lead, Machine Learning)

  1. What position do you want to have when you are done with you schooling?

  2. Write job descriptions for each position for starting a data science team for self driving cars.

Overview of class

My goal is to give you a one degree of distance to the stuff thats happening in the book.

There is a lot of technical syntax to consider in this class, my hope is our discussions will be more about the big overview nad how to integrate DS into work.

How to learn:

  • some algorithms are designed best run on silicon and others on carbon.
  • You need to develop ways of thinking about the problem that allows you to apply your techniques.
  • What question would you like to be able to answer with these skills that you can't currently?
  • Stack overflow life cycle (1. Can't find question, 2. Finding answers, 3. Realizing many answers are poor.)

What is machine learning

  • There is no simple solution to statistical validity. There is no function, but rather a conversation.
  • Four ways of thinking about data. Tall, flat, square, blocks

Language choice

  • coding is not about syntax Syntax is how you convey ideas, some ideas are specific to language, but most programming fundamentals are standard.  

  • python -> pandas

  • scripting versus programing, interactive environment versus output

  • Python packages versus R repo

Notes:

Keepers of the algorithms.

Magic trick.