W11 / 2026-11-16 / 20 MIN / ENGLISH

Optional midterm written and oral rehearsal

Optional written/oral exam rehearsal. This does not add a weekly graded assignment. The exercise time is included in class practice, not added after the three-hour meeting.

Open exercise in Colab Download exercise notebook Worked walkthrough All exercises

Before you start

Use a CPU runtime. Save a personal copy or open the downloaded notebook in Jupyter.
Run the included complete walkthrough first. Read its setup, definitions, and checkpoints.
Predict before running each exercise cell; then change only the requested factor.
Keep your actual outputs, configuration, explanation, and limitation. Free tools are sufficient.

TASK 1 / PREDICT - RUN - EXPLAIN

Explain precision versus recall

Use TP=8, FN=2, FP=3, TN=27. Write both denominators before running the calculation.

Write your prediction in the notebook response field before running.
Run the starter cell below after the worked walkthrough.
Make the requested controlled change and record the configuration.
Explain the mechanism using the actual output, not an invented result.

observations['task1']={'precision':8/(8+3),'recall':8/(8+2),'specificity':27/(27+3)}
print(observations['task1'])

Self-check: Recall is 0.8. Precision is 8/11 and specificity is 0.9.

TASK 2 / PREDICT - RUN - EXPLAIN

Find the leaked target

A target occurs at index 100. Compare these windows and explain the off-by-one error in a sentence.

Write your prediction in the notebook response field before running.
Run the starter cell below after the worked walkthrough.
Make the requested controlled change and record the configuration.
Explain the mechanism using the actual output, not an invented result.

good_history=np.arange(76,100); bad_history=np.arange(77,101)
observations['task2']={'good_last':int(good_history[-1]),'bad_last':int(bad_history[-1]),'bad_contains_target':100 in bad_history}
print(observations['task2'])

Self-check: The bad window includes index 100. A correct length does not establish a leakage-safe window.

TASK 3 / PREDICT - RUN - EXPLAIN

Give a two-minute oral defense

Without reading code, explain your task, data, baseline, split, result, and strongest limitation. Ask a partner what observation would falsify your claim.

Write your prediction in the notebook response field before running.
Run the starter cell below after the worked walkthrough.
Make the requested controlled change and record the configuration.
Explain the mechanism using the actual output, not an invented result.

observations['task3']={'oral_prompts':['What was predicted?','Which data were available?','What baseline was used?','What would falsify the claim?']}
print(observations['task3'])

Self-check: Use actual project evidence. This is optional practice, not an additional assessment or an answer key to the examination.

Completion checklist

Three predictions and three explanations.
One controlled change with its actual outcome.
One limitation and a brief AI-use/checking disclosure, or "none".
A downloaded W11-evidence.json, then a two-minute peer explain-back.

The notebook marks incomplete response records as drafts. Completing the fields is not proof that an explanation is correct.

The guided walkthroughs were previously executed. The new exercise extension cells have not yet been execution-tested for this revision. Agent exercises are deterministic simulations; no paid model or consequential external action is used.

Return to this week Exercise collection