justmarkham / python-data-science-workshop

Workshop: Python for Data Science

Python for Data Science

Workshop at General Assembly (Washington, DC) on November 23, 2014.

Instructors: Kevin Markham and Sinan Ozdemir

Agenda

1. Why Python for data science?
2. Python basics
   • Code
3. Getting data
4. pandas for data exploration and visualization
   • Alcohol consumption data: FiveThirtyEight article, their GitHub repository, modified data
   • Code
   • split-apply-combine pattern
5. scikit-learn for machine learning
   • Introduction to machine learning (slides)
   • Code
6. Recommended resources for self-learning
   • Basic Python: Codecademy, Google's Python Class, Python Tutor (to visualize code execution)
   • Pandas: tutorial, book: "Python for Data Analysis" (includes numpy and Python reference)
   • Web scraping: tutorial
   • Command line: tutorial
   • Git and GitHub: video series
   • Machine learning: book and videos: "An Introduction to Statistical Learning", scikit-learn tutorials, Data Science as Competitive Sport (video), Kaggle Titanic competition
   • Types of data scientists: "Analyzing the Analyzers" (ebook written by the founders of Data Community DC)
   • Full-fledged courses: Data Science Specialization (9 short courses by JHU in R), Machine Learning (1 course by Andrew Ng in Matlab/Octave), Learning from Data (1 course, programming language not specified)
7. General Assembly's Data Science course
   • Official page
   • Current repo
   • Project examples
8. Ask us anything!