DS Datasets

All datasets are student-visible and may be freely copied. Every dataset that contains defects documents them here, together with the expected values a correct program must produce — so students can self-check and graders compare.

Generated inputs are also welcome; when a task says "use the dataset", use these files so outputs are comparable.

Files

temperatures.txt — 18 clean readings

No defects. For DS-04.

• mean = 20.8 (exact sum 374.4 / 18 = 20.7999…, prints 20.8 at 1 dp)
• values strictly above the true mean: 6 (the six ≥ 21.8)

survey_ratings.txt — 40 non-blank lines

For DS-06. Defects on purpose:

• 35 valid ratings 1..5
• 3 out-of-range (0, 9, -2)
• 2 non-numeric (five, missing?)
• counts: 1→3, 2→4, 3→6, 4→14, 5→8; relative for 4 = 14/35 = 0.40

readings_na.txt — 15 non-blank lines (+ 1 blank line to ignore)

For DS-09. Line classes:

• 11 loaded (all parse and lie in −50..60)
• 2 missing (NA on N-02 and N-03)
• 2 invalid (twenty non-numeric; N-06 has no value field)
• class identity holds: 11 + 2 + 2 = 15
• mean of loaded = 21.0455… → 21.0 at 1 dp (231.5 / 11)

scores.csv — 11 lines

For DS-13. Defects on purpose:

• header name,score — per the task's stated rule, it classifies as bad-score (it is data-shaped, that is the lesson)
• Farah 90 — no comma → no-comma (1)
• Gus,101 (range) and Ivan,abc (non-numeric) → bad-score (2 more)
• 7 loaded: 88, 72, 95, 61, 79, 90, 100
• mean = 83.57 (585/7); max 100 owned by Jade (unique owner)

students.txt — 12 lines

For DS-14. Defects:

• Eli 300 2.0 — age out of 16..90 → rejected
• Gus 22 4.1 — gpa out of 0..4 → rejected
• Ivy 21 three — non-numeric gpa → rejected
• 9 valid rows; overall mean age 20.89, mean gpa 3.37
• Ana appears 3× (3.4, 3.9, 3.6) → Ana mean 3.63; Joy and Cy are single-row groups; first-appearance order: Ana, Ben, Cy, Dee, Fay, Gus, Hal, Joy

messy_grades.txt — 20 non-blank lines

For DS-15. Defect classes (your program must state its classification rule for the 3 no-name lines; suggested rule: no name ⇒ corrupt):

• 13 valid (min 41, max 100)
• 2 missing (Ben NA, Ned NA)
• 2 corrupt values (Dan abc, Qu 101 out of range)
• 3 no-name lines (72, 79, the lone NA)
• expected: mean 74.92, median 79.00, usable 65.0 % (13/20)

stations/ — three station files + index.txt

For DS-18. Format day temperature; temperature plausible range −50..60.

• hottest single reading: 27.5 at station_south day 7
• coldest single reading: 17.5 at station_coast day 4
• mean ranking (hottest first): south 26.07, north 21.06, coast 18.75
• no ties in this data; the tie rule (alphabetical) exists for robustness — test it by adding a temporary file with an equal mean.