Final Exam Model Paper 1

i. Each call runs the function, which prints its price, so the screen shows 320, then 480. But the function has no return, so each call gives back None, and line 4 tries to work out None + None, which Python can’t add:

320480

and then TypeError: unsupported operand type(s) for +: 'NoneType' and 'NoneType'. Printing a value shows it to a person; only return gives it back to the program.
ii. Return the price instead of printing it:

def plate_price(plates):    """The price in rupees of this many plates."""    return plates * 160 total = plate_price(2) + plate_price(3)print("Total:", total)

Total: 800

• i.It is joined up and has no loops: each match except the final sends its winner to exactly one next match, its parent, and no winner ever goes back to an earlier match. The root is the final, which has no parent; the leaves are the four first-round matches, which have no matches below them. (7 matches, 6 edges: a tree with n nodes has n − 1 edges.)
• ii.Every match has only one parent, so going up from a first-round match there is only one way: that match, its semi-final, the final. It is the list of matches a team starting there must win to lift the cup: three wins, in that order.

SELECT species, plantedFROM treesWHERE ward = 9 AND alive = 'no'ORDER BY planted DESC;

Dates written as year-month-day sort in the right order, so DESC puts the most recent first.

• i.Mean = (820 + 1,150 + 640 + 990 + 1,300 + 760 + 9,400) ÷ 7 = 15,060 ÷ 7 ≈ Rs 2,151. In order: 640, 760, 820, 990, 1,150, 1,300, 9,400, so the median is the middle one, Rs 990.
• ii.The median. The mill’s bill is an outlier, far above the rest, and it pulls the mean up to more than any other house pays. The median isn’t moved by how far out the outlier is.

• i.Python found and started quiz.py; it is the program that can’t find its file. open("questions.txt") names no folder, so it looks in the working directory, the folder Kiran ran it from, ~/school/class10. questions.txt is in data, one folder further in.
• ii.Any one: change line 1 to open("data/questions.txt"); or move questions.txt into class10; or run the program from the data folder: cd data, then python3 ../quiz.py.

i. cd school takes her into school, so mkdir club makes club inside school, beside class9 and class10. cd club goes into it; .. is the folder one level up, school, so cd ../class10 goes to class10, and pwd prints:

/home/pema/school/class10

ii. Up one level and into club:

cd ../club

The prompt then shows pema@lab-pc:~/school/club$. (cd ~/school/club or cd /home/pema/school/club works as well.)

• i.After pretraining, it was fine-tuned on example conversations written by people (a request and a good answer), and then trained on people’s comparisons of its answers, choosing the better one (feedback from people, RLHF).
• ii.Size and pretraining give a model its knowledge, but a base model only carries on text: given a question, it may write more questions. Fine-tuning and feedback teach it to answer a request helpfully, in the way people prefer. People judge the answers they get, so a smaller model that answers well beats a bigger one that doesn’t answer at all.

Kiran’s. His questions were chosen before he saw any answers, so he couldn’t pick them, even without meaning to, to make the chatbot look good or bad; every answer was checked in a reliable source; 60 questions are far more than 5, so one lucky or unlucky answer matters less; and counting by group can show whether it is worse about Karnali than about Bagmati, which an overall score would hide.

Any two, for example:

• Software developer: writes, tests and fixes programs and apps.
• Data analyst: cleans data, finds patterns and makes honest charts for decisions.
• Network engineer: designs and runs the networks of internet companies, banks and offices.
• Security analyst: finds weaknesses before attackers do, and responds to attacks.
• UI/UX designer: designs screens and how people use them, and tests them with users.
• IT support or system administrator: keeps a school’s, hospital’s or company’s computers and accounts running.
• GIS and mapping specialist: makes and analyzes maps from satellite and GPS data.
• Machine learning engineer: builds and tests AI models.
• Computing teacher: teaches computing to the next people.

It leaves out people without a smartphone or a connection (many older people, poorer households, villages with a weak signal) and people who don’t read English well: an inequity in access, made by design choices. Changes that let them in: keep a counter at the ward office, or let people pay at a bank; make the app work in Nepali; or let a family member or a ward helper pay for them.

i. and ii.

"""The homestay's charges.""" PER_NIGHT = 1500     # rupees, for each guestPER_MEAL = 350       # rupees  def bill(guests, nights, meals):    """The bill in rupees: the nights, 10% off for 3 nights or more, plus the meals."""    stay = guests * nights * PER_NIGHT    if nights >= 3:        stay = stay * 0.9    return round(stay + meals * PER_MEAL)  if __name__ == "__main__":    # tests: 2 nights get no discount, 3 nights do    print(bill(1, 2, 0), "should be 3000")    print(bill(1, 3, 0), "should be 4050")    print(bill(2, 1, 2), "should be 3700")

iii.

import homestay print("Bill: Rs", homestay.bill(4, 3, 24))

Bill: Rs 24600

4 guests × 3 nights × Rs 1,500 = Rs 18,000, less 10% = Rs 16,200; 24 meals × Rs 350 = Rs 8,400; together Rs 24,600. Any clear names and messages are fine.

i.

• ii.3 steps. Hemant was first reached from Esha, Esha from Bikram, Bikram from Aarati: Aarati → Bikram → Esha → Hemant.
• iii.Esha can now be reached only through Hemant, who is 3 steps away by Dipa and Gopal: 4 steps, Aarati → Dipa → Gopal → Hemant → Esha.

• i.The two players of Makalu: Dawa Sherpa and Firoj Ansari.
• ii.One row per player, with the number of goals, most first, and only the first two kept: 3, 3 (Chirag Shah scored 3) and 2, 2 (Binita Magar scored 2). Everyone else scored 1.

iii. Each goal of match 2 joined to its scorer’s row, earliest first:

iv. So that each fact is kept once. A player’s name and house are written once, in players, and each goal points to that row with a foreign key. Writing the name and house in every goal would repeat them (Chirag Shah three times), and if one copy were changed or misspelled, the copies would disagree; the ID also tells apart two players with the same name.

• i.str (text). pandas makes a column numbers only if every value reads as a number, and - and 14 L don’t.
• ii.It joins the strings end to end, as + does for text:

1291511-14 L

That is not a sum: adding text sticks it together. Nothing warns you, which is why you look at dtypes before working with a column.

• iii.Left missing. “-” means nobody knows the amount, not that she brought none; writing 0 would pull Panauti’s average and median down with a number nobody measured. (If the clerk is asked and she really brought none, then 0 is right.)
• iv.Each village’s middle value, the missing one left out:

villageDhulikhel    14.0Panauti      10.5Name: liters, dtype: float64

Dhulikhel: 11, 14, 15, so 14. Panauti: 9 and 12 (Gita’s gap is skipped), so the middle is (9 + 12) ÷ 2 = 10.5.

• i.Where the satellite is at that moment, and the exact time the signal left it.
• ii.GPS only needs the phone to listen. It notes when each signal arrives, works out how long it took, and turns the time into a distance (speed of light × travel time). From its distances to several satellites (four, one for its own clock’s error) it works out where it is: trilateration. No mobile network is used.
• iii.Because of Einstein’s relativity: up there gravity is weaker, which makes the satellites’ clocks run fast, and their speed makes them run a little slow; together they would gain about 38 microseconds a day on clocks on the ground. Left alone, that would move every position by about 10 km a day; ticking slightly slow cancels it.
• iv.With mobile data the phone downloads where the satellites are (assisted GPS), so it doesn’t have to wait to hear that from the satellites’ own slow signals.

"""The ward library's late fees.""" MAX_FEE = 200    # rupees, for one book  def late_fee(days):    """The fee in rupees for a book returned this many days late."""    if days <= 7:        fee = days * 5    else:        fee = 7 * 5 + (days - 7) * 10    return min(fee, MAX_FEE)  if __name__ == "__main__":    # tests at the edges: on time, the 7th and 8th days, and the cap    print(late_fee(0), "should be 0")    print(late_fee(7), "should be 35")    print(late_fee(8), "should be 45")    print(late_fee(30), "should be 200")

Run by itself, library.py prints its tests:

0 should be 035 should be 3545 should be 45200 should be 200

import library total = 0for book in range(1, 4):    days = int(input(f"Book {book}: how many days late? "))    total = total + library.late_fee(days)print("Total to pay: Rs", total)

Book 1: how many days late? 4Book 2: how many days late? 9Book 3: how many days late? 45Total to pay: Rs 275

4 days: Rs 20. 9 days: Rs 35 + 2 × 10 = Rs 55. 45 days: Rs 35 + 38 × 10 = Rs 415, capped at Rs 200. Together Rs 275.

The same program, with these three lines in place of the last one:

import sqlite3 db = sqlite3.connect(":memory:")db.execute("CREATE TABLE birds (bird_id INTEGER PRIMARY KEY, name TEXT, kind TEXT)")db.execute("""CREATE TABLE sightings (sighting_id INTEGER PRIMARY KEY,              bird_id INTEGER REFERENCES birds(bird_id),              place TEXT, day TEXT, how_many INTEGER)""")db.executemany("INSERT INTO birds VALUES (?, ?, ?)", [    (1, "Spiny Babbler", "resident"),    (2, "Red-vented Bulbul", "resident"),    (3, "Common Myna", "resident"),    (4, "Ruddy Shelduck", "winter visitor"),    (5, "Great Barbet", "resident"),    (6, "Oriental Magpie-Robin", "resident"),])db.executemany("INSERT INTO sightings VALUES (?, ?, ?, ?, ?)", [    (1, 2, "Phulchowki", "2026-01-10", 4),    (2, 5, "Phulchowki", "2026-01-10", 2),    (3, 1, "Phulchowki", "2026-01-10", 1),    (4, 4, "Taudaha", "2026-01-11", 23),    (5, 3, "Taudaha", "2026-01-11", 6),    (6, 2, "Taudaha", "2026-01-11", 3),    (7, 6, "Shivapuri", "2026-01-12", 2),    (8, 5, "Shivapuri", "2026-01-12", 3),    (9, 2, "Shivapuri", "2026-01-12", 5),    (10, 1, "Shivapuri", "2026-01-12", 1),])  def ask(sql):    """Print each row the query gives, then a blank line."""    for row in db.execute(sql):        print(row)    print()  ask("SELECT bird_id, how_many FROM sightings WHERE place = 'Shivapuri' ORDER BY how_many DESC")ask("SELECT place, SUM(how_many) FROM sightings GROUP BY place")ask("""SELECT birds.name, sightings.how_many FROM sightings       JOIN birds ON sightings.bird_id = birds.bird_id       WHERE sightings.place = 'Taudaha' ORDER BY sightings.how_many DESC""")

(2, 5)(5, 3)(6, 2)(1, 1) ('Phulchowki', 7)('Shivapuri', 11)('Taudaha', 32) ('Ruddy Shelduck', 23)('Common Myna', 6)('Red-vented Bulbul', 3)

(The counts are made up for this paper.)

school = {    "Gate": ["Playground", "Office"],    "Playground": ["Gate", "Hall", "Canteen", "Garden"],    "Office": ["Gate", "Hall"],    "Hall": ["Office", "Playground", "Library", "Computer lab"],    "Library": ["Hall", "Science lab"],    "Computer lab": ["Hall", "Science lab"],    "Science lab": ["Library", "Computer lab"],    "Canteen": ["Playground", "Garden"],    "Garden": ["Playground", "Canteen"],    "Store room": [],}  def fewest_steps(graph, start, goal):    """The route from start to goal with the fewest steps, or None."""    came_from = {start: None}    to_visit = [start]    while len(to_visit) > 0:        place = to_visit.pop(0)        if place == goal:            route = []            while place is not None:                route.append(place)                place = came_from[place]            route.reverse()            return route        for nxt in graph[place]:            if nxt not in came_from:                came_from[nxt] = place                to_visit.append(nxt)    return None  for start, goal in [("Science lab", "Gate"), ("Canteen", "Computer lab"), ("Store room", "Gate")]:    route = fewest_steps(school, start, goal)    if route is None:        print(start, "to", goal + ": No way out")    else:        print(start, "to", goal + ":", " → ".join(route), "(" + str(len(route) - 1), "doorways)")

Science lab to Gate: Science lab → Library → Hall → Office → Gate (4 doorways)Canteen to Computer lab: Canteen → Playground → Hall → Computer lab (3 doorways)Store room to Gate: No way out

Through the Computer lab, or through the Playground instead of the Office, is also 4 doorways: any route of 4 is right.

• i.The rule changes between the 7th and the 8th day, and mistakes hide at the edges (< instead of <=, an off-by-one in the days after the first week), so testing both sides of the edge catches them. Without the if, the tests would run every time main.py imports library, printing their lines in the middle of the member’s bill.
• ii.The bird’s name is in the birds table, but the place is in sightings: the JOIN puts each sighting beside its bird. ON says which rows go together: a sighting’s bird_id (the foreign key) must equal the bird’s bird_id (its primary key).
• iii.pop() takes the newest place from the back: the list becomes a stack, and the search dives down one path first (depth-first). It still finds a way out if there is one, but not always the one with the fewest doorways, because it no longer searches in rings.