Skip to content

Latest commit

 

History

History
113 lines (81 loc) · 3.41 KB

README.md

File metadata and controls

113 lines (81 loc) · 3.41 KB

Python for Engineering

This class focuses on (1) exposure to concepts and (2) providing templates for success.

The underlying goal is to achieve good taste in software, which can be reached much earlier than complete understanding.

The repetition required to fully learn the material is expected to happen under self-study or on-the-job outside class time.

Syllabus

This class was originally taught in one hour per week segments over the course of 10 weeks without a rigidly fixed schedule.

Multiple groups were run simultaneously, and while schedule was largely driven by discussion, the variability in class time between the groups was only about 10%.

  • Setup & git version control primer
  • Language fundamentals - imports, syntax, types, collections, patterns
  • Base libraries
  • Classes - definitions, inheritance, methods, properties, caching
  • Packaging - project structure, build systems, versioning, executable entrypoints, dependency management, github repo configuration, documentation, command line interfaces, logging
  • Quality control - testing (pytest, doctesting, example testing, coverage), linting & formatting, type hinting & checking, CI w/ github actions, publishing

Installation & Setup

Install essentials

  • If you're on Windows, install WSL2
    • Anaconda-managed python environments are not recommended for this project
  • Python (3.11 or newer)
    • Preferably via uv, which allows managing multiple pythons
  • Pip, uv, git, vscode

Set up a virtual environment

  • This allows deconflicting dependencies between projects and (usually) prevents you from accidentally bricking your system env
  • Do not work in your global environment! It will rapidly become misconfigured

Using <stuff> to denote examples and placeholders.

cd ~/
mkdir envs
cd envs
uv venv <short_env_name_without_spaces_like_"310">
ls # should show a folder called `310` or whatever you named it

Then, to add a convenience function to activate that virtual env:

nano ~/.bashrc # linux
OR
nano ~/.profile # mac

At the bottom of the file, add this function:

activate () {
  source ~/envs/$1/bin/activate ;
 }

This function will be available next time you open a new terminal, or after running source ~/.bashrc (linux) or source ~/.profile (mac).

Activate the virtual env using the new function like activate 310. Now, any python packages you install will be placed in that virtual env instead of overwriting your system env.

Install python deps in virtual environment

Install the class's python library deps:

uv pip install -r requirements.txt

Git Reference

Using <stuff> to denote examples and placeholders.

# Copy a repo to a local folder
git clone <repo clone link>

# Check what has been modified or staged
git status

# Update local version from server
git pull

# Make a branch
git checkout -b <branch name>

# Switch to a branch that already exists
git checkout <branch name>

# Send changes to server
#   `add` "stages" changes locally; Marks things to commit
#   `commit` marks a group of changes; still locally
#   `push` sends changes to the server
git add .  
git commit -m "<some message>"
git push

# Obliterate local changes (only for files already controlled by the repo)
git reset --hard

# Delete local files and folders that aren't version-controlled
git clean -fd

Git has a lot more features not discussed here that are best found when needed.