💻 Work environment and submission workflow

In addition to theory 📖 this course has a strong ⚙️ practical component, with hands-on exercises and a semester-long project. To follow the course effectively you need a local development environment to interact with the course materials, and to write and run code.

This class also sets up the workflow you will use to submit every homework: a pull request to the course repository.

Prerequisites¶

• You have a fairly modern computer: desktop or laptop, Windows or Mac or Linux

• You have administrative access to this computer (can install programs)

Python and Anaconda¶

The Anaconda distribution is an easy way to install:

• Python

• Jupyter Notebook

• Scientific programming libraries

Anaconda Distribution

The course uses numpy, scipy, matplotlib and sympy. All of these ship with Anaconda; no other frameworks are required.

Git and Git GUI¶

• Git is the command line version control software

• A GUI makes Git a lot more practical

Git · Git GUI applications

Recommended options:

• GitHub Desktop

• SourceTree

• VS Code editor (has Git built in)

GitHub¶

• Social coding network website

  • hosting code

  • version control and integrations

  • community of coders and open source projects

  • benefits for students

  • free hosting for static web pages — this book is one

• Register on GitHub if you have not yet

  • be mindful about using your personal data when registering

  • apply for the Student Developer Pack

• Course materials and all homework are distributed and collected through the course GitHub repository

Text editor¶

• A good text editor is invaluable for editing source files

• Editing could be done in Jupyter or other default editors, but it is less convenient

• Essential for bigger coding projects — such as your course project

• Good options are:

  • VS Code

  • Sublime Text (paid)

  • PyCharm (full IDE)

Jupyter notebooks¶

• An excellent way to present and discuss code

  • this entire course is taught using notebooks

• A good instrument to develop new ideas

  • especially together with coauthors

• Saved to disk as JSON files with multiple sections

  • text

  • math and formulas in LaTeX

  • code (different languages are possible)

  • output from the code

Jupyter notebooks have limitations

• NOT a good way to store developed code — use libraries (modules)

• NOT good for version control

  • changes in metadata are tracked

  • changes in output are tracked

  • merging changed files may break the JSON format

• Require additional tools to work well with Git

• We don’t worry about this for most of the course, but you should for your project

Homework submission workflow¶

Every homework in this course is submitted as a pull request to the course repository. The mechanics are the same every time:

1. Fork the course repository to your own GitHub account (once, at the start of the semester)

2. Clone your fork to your machine git clone <address copied from the GitHub page>

3. Branch for each assignment: git checkout -b hw1-yourname

4. Work in your own folder — submissions/yourname/hw1/ — so that submissions never collide with each other

5. Commit with a meaningful message: git add . && git commit -m "HW1: inventory model solver"

6. Push the branch to your fork: git push origin hw1-yourname

7. Open a pull request against the course repository before the deadline

Homework is discussed at the start of the class that follows it, with one student presenting the solution at the board. The presenter rotates, so plan on presenting several times over the semester — including code that does not work yet, which is usually the more instructive case.