Records, split and early activity
Task 7 · worked step 1
The supplied split tells you which records teach the rules and which records only receive predictions. Keep those roles separate.
Lab 4 PDF pp. 6–7 · Task 7
Your task
Use S1–S4 to learn all parameters and the prediction rule. Use T1–T2 only for held-out predictions and later scoring. Predict N1 separately; its true outcome is unknown.
Evidence to include
Training learns. Test checks. New waits.
Worked example and code
Try the task in the PDF first, then compare your working below. Run the code in your own Python notebook.
Open Lab 4 instructions (PDF)| ID/set | Score; minutes | W1-W4 total logins; programme; outcome |
|---|---|---|
| S1/train | 60; 60 | 2; CS; Withdraw |
| S2/train | 65; 90 | 4; Economics; Withdraw |
| S3/train | 80; 180 | 12; Psychology; Stay |
| S4/train | 85; missing | 16; CS; Stay |
| T1/test | 62; 70 | 2; CS; Withdraw |
| T2/test | 82; 170 | 14; Economics; Stay |
| N1/new | 78; missing | 10; Physics; unknown |
All these early features are stated to be available at the cutoff. The four S rows fit the rules. T outcomes are used only after predictions; N1's true outcome remains unknown.
Practice problems
Supplementary study questions on the lecture concepts — not verbatim exam questions. Each answer counts toward spaced review.
1. Conceptual understanding
Definitions and distinctions.
Which rows fit the combined Lab 4 rules?
Can N1 be included in test accuracy?
2. Application
Apply the concept to a case or calculation.
Which rows are held out for evaluation in the combined lab?
How many students fit the combined lab's parameters?
3. Interpretation and limitations
Evaluate assumptions and justify conclusions.
A student has weekly logins [1,2,0,3] in weeks 1–4. What is early_logins?
Why exclude N1 from test error?
Explain it in writing
Self-checked, not marked. Write a few sentences, then compare with the example answer. Your writing is saved as you type.
Show an example answer
They are held-out outcomes. Predictions must be made from the training rule first, then compared with these labels for evaluation.
Check: did you answer the question, use evidence or working, and avoid claiming more than the evidence supports?
Show an example answer
Report its predicted label separately. Without its true outcome, do not include N1 in the error or accuracy denominator.
Check: did you answer the question, use evidence or working, and avoid claiming more than the evidence supports?
Saved only in this browser. This note doesn’t fill in the downloaded notebook; “reviewed” is a personal checklist, not a grade.