Datasets:

Asilarkness
/

DiffuRefill-SFT

+{"tid": "e8f1c0f070", "cat": "finance", "code": "def make(rng, original=False):\n    if original:\n        income = 120000\n        tax_rate = 25\n        payout = 60\n        tax = income * tax_rate // 100\n        after_tax = income - tax\n        dividends = after_tax * payout // 100\n        question = (\"A partnership between Himan and Heeb generates an annual pre-tax net income of $120,000. \"\n                    \"The corporate tax rate is 25%, and the partners agree to distribute 60% of the after-tax \"\n                    \"net income to shareholders as dividends. What is the total dollar amount of dividends paid out for the year?\")\n        lines = [\n            f\"Pre-tax net income = ${income:,}\",\n            f\"Corporate tax = ${income:,} \u00d7 {tax_rate / 100:.2f} = ${tax:,}\",\n            f\"After-tax net income = ${income:,} \u2212 ${tax:,} = ${after_tax:,}\",\n            f\"Dividend payout ratio = {payout}%\",\n            f\"Dividends paid = ${after_tax:,} \u00d7 {payout / 100:.2f} = ${dividends:,}\",\n        ]\n        return question, lines, f\"${dividends:,}\"\n\n    names = [\"Ava\", \"Ben\", \"Cara\", \"Dana\", \"Eli\", \"Fay\", \"Gus\", \"Hana\", \"Ivan\", \"Jade\",\n             \"Kai\", \"Lena\", \"Mona\", \"Nico\", \"Omar\", \"Pia\", \"Rex\", \"Sara\", \"Tara\", \"Uli\"]\n    valid_incomes = [v * 1000 for v in range(10, 101) if v != 120]\n    valid_tax_rates = [v for v in range(1, 41) if v != 25]\n    valid_payouts = [v * 10 for v in range(1, 10) if v != 6]\n    templates = [\n        \"A partnership between {n1} and {n2} operates with {f1}, {f2}, and {f3}. What is the total dollar amount of dividends paid out for the year?\",\n        \"A partnership between {n1} and {n2} reports {f1}, {f2}, and {f3}. What is the total dollar amount of dividends paid out for the year?\",\n        \"A partnership between {n1} and {n2} has {f1}, {f2}, and {f3}. What total dollar amount of dividends is paid out for the year?\",\n    ]\n\n    for _ in range(50):\n        income = rng.choice(valid_incomes)\n        tax_rate = rng.choice(valid_tax_rates)\n        payout = rng.choice(valid_payouts)\n\n        if (income * tax_rate) % 100 != 0:\n            continue\n        tax = income * tax_rate // 100\n        after_tax = income - tax\n        if after_tax <= 0:\n            continue\n        if (after_tax * payout) % 100 != 0:\n            continue\n        dividends = after_tax * payout // 100\n        if dividends <= 0:\n            continue\n\n        n1, n2 = rng.sample(names, 2)\n        facts = [\n            f\"an annual pre-tax net income of ${income:,}\",\n            f\"a corporate tax rate of {tax_rate}%\",\n            f\"an agreement to distribute {payout}% of the after-tax net income to shareholders as dividends\",\n        ]\n        rng.shuffle(facts)\n        template = rng.choice(templates)\n        question = template.format(n1=n1, n2=n2, f1=facts[0], f2=facts[1], f3=facts[2])\n\n        lines = [\n            f\"Pre-tax net income = ${income:,}\",\n            f\"Corporate tax = ${income:,} \u00d7 {tax_rate / 100:.2f} = ${tax:,}\",\n            f\"After-tax net income = ${income:,} \u2212 ${tax:,} = ${after_tax:,}\",\n            f\"Dividend payout ratio = {payout}%\",\n            f\"Dividends paid = ${after_tax:,} \u00d7 {payout / 100:.2f} = ${dividends:,}\",\n        ]\n        return question, lines, f\"${dividends:,}\"\n\n    return None\n"}
+{"tid": "15e7207ae6", "cat": "probability", "code": "import math\n\ndef make(rng, original=False):\n    def plural(n, one, many):\n        return f\"{n} {one if n == 1 else many}\"\n\n    if original:\n        question = (\n            \"A film festival committee selects a jury of 5 members from a pool of 12 candidates, \"\n            \"consisting of 7 women and 5 men. What is the probability that the jury will consist \"\n            \"of exactly 3 women and 2 men? Express your answer as a common fraction in simplest form.\"\n        )\n        w, m, n, k, tw = 7, 5, 12, 5, 3\n        tm = k - tw\n        cw = math.comb(w, tw)\n        cm = math.comb(m, tm)\n        fav = cw * cm\n        tot = math.comb(n, k)\n        lines = [\n            f\"C({w},{tw}) = {cw}\",\n            f\"C({m},{tm}) = {cm}\",\n            f\"{cw} * {cm} = {fav}\",\n            f\"C({n},{k}) = {tot}\",\n            f\"{fav}/{tot}\",\n            f\"{fav}/2 = {fav // 2}\",\n            f\"{tot}/2 = {tot // 2}\",\n            f\"gcd({fav // 2}, {tot // 2}) = {math.gcd(fav // 2, tot // 2)}\",\n            f\"{fav // 2}/{tot // 2}\",\n        ]\n        answer = f\"\\\\frac{{{fav // 2}}}{{{tot // 2}}}\"\n        return question, lines, answer\n\n    subjects = [\n        \"A film festival committee\",\n        \"A science fair board\",\n        \"A debate competition\",\n        \"A book club\",\n        \"A conference committee\",\n        \"A sports league\",\n    ]\n    groups = [\"jury\", \"panel\"]\n\n    for _ in range(100):\n        n = rng.randint(10, 18)\n        w = rng.randint(max(2, n // 3), min(n - 2, n - n // 3))\n        m = n - w\n        k = rng.randint(3, min(7, n - 3))\n\n        low = max(1, k - (m - 1))\n        high = min(w - 1, k - 1)\n        if low > high:\n            continue\n\n        tw = rng.randint(low, high)\n        tm = k - tw\n\n        if (n, w, m, k, tw, tm) == (12, 7, 5, 5, 3, 2):\n            continue\n\n        cw = math.comb(w, tw)\n        cm = math.comb(m, tm)\n        fav = cw * cm\n        tot = math.comb(n, k)\n\n        if fav == 0 or fav == tot:\n            continue\n\n        subject = rng.choice(subjects)\n        group = rng.choice(groups)\n        question = (\n            f\"{subject} selects a {group} of {k} members from a pool of {n} candidates, \"\n            f\"consisting of {plural(w, 'woman', 'women')} and {plural(m, 'man', 'men')}. \"\n            f\"What is the probability that the {group} will consist of exactly \"\n            f\"{plural(tw, 'woman', 'women')} and {plural(tm, 'man', 'men')}? \"\n            \"Express your answer as a common fraction in simplest form.\"\n        )\n\n        lines = [\n            f\"C({w},{tw}) = {cw}\",\n            f\"C({m},{tm}) = {cm}\",\n            f\"{cw} * {cm} = {fav}\",\n            f\"C({n},{k}) = {tot}\",\n            f\"{fav}/{tot}\",\n        ]\n\n        g = math.gcd(fav, tot)\n        num = fav // g\n        den = tot // g\n\n        if g > 1:\n            lines.append(f\"gcd({fav}, {tot}) = {g}\")\n            lines.append(f\"{fav}/{g} = {num}\")\n            lines.append(f\"{tot}/{g} = {den}\")\n        else:\n            lines.append(f\"gcd({fav}, {tot}) = 1\")\n\n        lines.append(f\"{num}/{den}\")\n        answer = f\"\\\\frac{{{num}}}{{{den}}}\"\n        return question, lines, answer\n\n    return None\n"}
+{"tid": "f2afa70d4a", "cat": "wordproblem", "code": "def make(rng, original=False):\n    def noun(n, plural, singular):\n        return singular if n == 1 else plural\n\n    def period(n):\n        return \"1 hour\" if n == 1 else f\"{n} hours\"\n\n    if original:\n        total_items = 60\n        xr, yr, zr = 2, 3, 1\n        h = 4\n        yh = h - 1\n        x = xr * h\n        y = yr * yh\n        z = zr * h\n        total = x + y + z\n\n        question = (\n            f\"Three artisans, X, Y, and Z, are repairing a set of {total_items} historic lampi\u00f5es. \"\n            f\"X can repair {xr} lampi\u00f5es per hour, Y can repair {yr} per hour, and Z can repair {zr} per hour. \"\n            f\"They work together for {h} hours, but Y stops working for the last hour due to a break. \"\n            f\"How many lampi\u00f5es have been repaired by the end of the {h}-hour period?\"\n        )\n\n        lines = [\n            f\"X repairs {xr} lampi\u00f5es per hour for {h} hours, totaling {x} lampi\u00f5es.\",\n            f\"Y repairs {yr} lampi\u00f5es per hour for {yh} hours, totaling {y} lampi\u00f5es.\",\n            f\"Z repairs {zr} lampi\u00e3o per hour for {h} hours, totaling {z} lampi\u00f5es.\",\n            f\"The total number of lampi\u00f5es repaired is {x} + {y} + {z} = {total}.\"\n        ]\n\n        return question, lines, str(total)\n\n    names = rng.sample([\"A\", \"B\", \"C\", \"D\", \"E\", \"F\", \"G\", \"H\"], 3)\n    rates = [rng.randint(1, 10) for _ in range(3)]\n    hours = rng.randint(2, 8)\n    stopper = rng.randrange(3)\n\n    if len(set(rates)) < 2:\n        return None\n\n    amounts = [rates[i] * (hours if i != stopper else hours - 1) for i in range(3)]\n    total_repaired = sum(amounts)\n\n    if total_repaired <= 0:\n        return None\n\n    total_items = rng.randint(total_repaired + 1, total_repaired + 60)\n    if total_items <= total_repaired:\n        return None\n\n    obj, sing = rng.choice([\n        (\"lamps\", \"lamp\"),\n        (\"clocks\", \"clock\"),\n        (\"vases\", \"vase\"),\n        (\"chairs\", \"chair\"),\n        (\"tables\", \"table\"),\n        (\"tools\", \"tool\")\n    ])\n\n    rate_clue = (\n        f\"{names[0]} can repair {rates[0]} {noun(rates[0], obj, sing)} per hour, \"\n        f\"{names[1]} can repair {rates[1]} {noun(rates[1], obj, sing)} per hour, \"\n        f\"and {names[2]} can repair {rates[2]} {noun(rates[2], obj, sing)} per hour.\"\n    )\n\n    time_templates = [\n        \"They work together for {hours} hours, but {stop} stops working for the last hour due to a break.\",\n        \"They are scheduled to work for {hours} hours, but during the final hour {stop} is on break.\",\n        \"They are scheduled to work for {hours} hours; in the last hour, {stop} is on break, so only the other two work.\"\n    ]\n    time_clue = rng.choice(time_templates).format(hours=hours, stop=names[stopper])\n\n    if rng.random() < 0.5:\n        clues = [rate_clue, time_clue]\n    else:\n        clues = [time_clue, rate_clue]\n\n    question = (\n        f\"Three artisans, {names[0]}, {names[1]}, and {names[2]}, are repairing a set of {total_items} historic {obj}. \"\n        f\"{clues[0]} {clues[1]} How many {obj} have been repaired by the end of the {hours}-hour period?\"\n    )\n\n    lines = []\n    for i, name in enumerate(names):\n        worked = hours if i != stopper else hours - 1\n        amount = amounts[i]\n        lines.append(\n            f\"{name} repairs {rates[i]} {noun(rates[i], obj, sing)} per hour for {period(worked)}, \"\n            f\"totaling {amount} {noun(amount, obj, sing)}.\"\n        )\n\n    lines.append(\n        f\"The total number of {obj} repaired is {amounts[0]} + {amounts[1]} + {amounts[2]} = {total_repaired}.\"\n    )\n\n    return question, lines, str(total_repaired)\n"}