CMSC 2208 Course Schedule
Course rhythm
Weekly module opens: Monday 8:00 AM
Assignments due: Sunday 11:30 PM
Work cadence:
Mon–Thu: work through reading + practice notebook
Sun: graded items due (when assigned)
Spring break: Mar 9–13, 2026
Final project block: Apr 27–May 10, 2026
Final project due: May 11, 2026
Schedule
| Week | Topics (with chapter/section anchors) | Deliverable | Reading (Müller & Guido) |
|---|---|---|---|
| 1 (Jan 12–Jan 18) | Course onboarding + “What is ML?” + environment setup + notebook workflow | Week 1 Verification Assignment (setup + screenshots + video) + D2L Quiz 1 | Ch. 1 |
| 2 (Jan 19–Jan 25) | First supervised model + scikit-learn workflow: kNN | D2L Quiz 2 | Ch. 2.1.1 (kNN) |
| 3 (Jan 26–Feb 01) | Linear models: classification + regression | D2L Quiz 3 | Ch. 2.2 (Linear Models) |
| 4 (Feb 02–Feb 08) | Trees + ensembles: random forests / gradient boosting | Skill Check 1 (autograded): train + report results on game stats + D2L Quiz 4 | Ch. 2.3 (Decision Trees) + 2.3.2 (Ensembles) |
| 5 (Feb 09–Feb 15) | Classifier outputs and confidence: decision function vs predicted probabilities | Reflection 1 (video): “Using uncertainty to compare models” + D2L Quiz 5 | Ch. 2.4: Uncertainty Estimates from Classifiers |
| 6 (Feb 16–Feb 22) | Preprocessing & scaling workflow: transformations and applying them correctly | Practice Notebook 3 (completion): “Scale game stats correctly (train vs test)” + D2L Quiz 6 | Ch. 3.3.1–3.3.4 |
| 7 (Feb 23–Mar 01) | Dimensionality reduction + visualization: PCA and t-SNE (“player map”) | Practice Notebook 4 (completion): “Build a player map (PCA / t-SNE)” + D2L Quiz 7 | Ch. 3.4.1 (PCA) + 3.4.3 (t-SNE) (NMF optional) |
| 8 (Mar 02–Mar 08) | Clustering: k-means, agglomerative, DBSCAN, and how to compare clusters | Practice Notebook 5 (completion): “Cluster players by playstyle” + D2L Quiz 8 | Ch. 3.5.1–3.5.5 |
| 9 (Mar 09–Mar 15) | Spring Break (no due dates) | — | — |
| 10 (Mar 16–Mar 22) | Representing data + feature engineering (one-hot, binning, interactions) | Practice Notebook 6 (completion): “Engineer features for matchmaking” + D2L Quiz 9 | Ch. 4 |
| 11 (Mar 23–Mar 29) | Model evaluation: train/test vs cross-validation + metrics beyond accuracy | Practice Notebook 7 (completion): “Evaluate models beyond accuracy” + D2L Quiz 10 | Ch. 5 (evaluation + metrics) |
| 12 (Mar 30–Apr 05) | Model selection + tuning: grid search + CV workflow | Skill Check 2 (autograded): metrics + CV + grid search summary + D2L Quiz 11 | Ch. 5 (model selection / grid search) |
| 13 (Apr 06–Apr 12) | Pipelines: preprocessing + model as one reproducible workflow | Practice Notebook 8 (completion): “Build a full pipeline for prediction” + D2L Quiz 12 | Ch. 6 |
| 14 (Apr 13–Apr 19) | Pipelines + tuning together (grid search inside pipeline; avoid leakage) | Skill Check 3 (autograded): pipeline + grid search + report best parameters + D2L Quiz 13 | Ch. 6 |
| 15 (Apr 20–Apr 26) | Text data intro (chat/reviews) or capstone prep emphasis | Reflection 2 / Project Proposal (graded): dataset + goal + metric + plan + D2L Quiz 14 | Ch. 7 (+ Ch. 8 possible) |
| 16 (Apr 27–May 03) | Final project work time (milestone: clean data + baseline + evaluation plan) | — (work week) | As needed (review Ch. 4–6) |
| 17 (May 04–May 10) | Final project work time (milestone: tuned model + results + narrative + video) | — (work week) | As needed |
Final project (Apr 27–May 11)
Due May 11, 2026: notebook report (pipeline + evaluation + interpretation) + short video walkthrough.