Detecting Anomalies in Smart Home Device Logs

A hands-on data science course (Python)

Who this is for: a teenager who can already write basic Python (variables, loops, functions) and wants a real project, not just tutorials.

How it works: each module ends with questions the student answers in their own words and code they write themselves. Nothing is pre-decided for them — they explore the real log, pick which anomalies matter, and build the pipeline. Answers live in each module's own answers.md (see Instructions below).

The data: a real exported device event log from an INSTEON/ISY-style home automation system — 116,407 rows, 121 devices, spanning May 2024 to Aug 2026. Not simulated. This means there's no pre-made answer key telling you which rows are anomalies — you have to build judgment for that yourself, which is closer to what real data science work is actually like.

Tools: Python 3, pandas, numpy, matplotlib. Later: scikit-learn. All free, all local.

---

What's here

• modules/ — the course itself, one folder per module (1-Setup/ through 10-Capstone/). Each folder holds that module's doc, any artifacts it needs (data files, skeleton .py code — e.g. modules/4-Data/devices_log.tsv, the real device event log, see modules/4-Data/description.md for column details), and your answers.md.

Instructions

Work through modules/ in order — don't skip ahead. For each module:

1. Finish the module. Read its doc (e.g. modules/1-Setup/setup.md), do the exercises, and follow along in your own code (a notebook, or the module's own skeleton .py file, as the module says).
2. Write it up in that module's answers.md. E.g. after modules/1-Setup/setup.md, fill in modules/1-Setup/answers.md, and record:
   • Answers — your response to every question the module asks, in your own words.
   • Thoughts — anything you noticed, found surprising, or want to remember.
   • Questions — anything that's still unclear or you'd want to ask about.
   • Completions — what you actually did/ran/built (files touched, commands run, what worked or didn't).
3. Commit it. git add modules/<N-Name>/answers.md and commit before moving to the next module — that write-up is what makes your progress checkable later.

A blank or missing answers.md for a module means that module isn't done yet, regardless of what code exists elsewhere.

Where to start

You don't have this repo cloned yet, so start by reading modules/1-Setup/setup.md in your browser — on GitHub, wherever this link was shared with you. Setup walks you through installing Python, Git, and VS Code, and cloning this repo (Step 7) — none of that can happen inside VS Code, because you don't have VS Code yet.

Once you've cloned the repo and opened it in VS Code (the second half of Step 7 onward), switch over to VS Code for everything else: finish the rest of Setup, write your answers into modules/1-Setup/answers.md, and commit and push it from VS Code's integrated terminal (git add, git commit, git push). Every module after that — reading, coding, answering, committing — happens inside VS Code.