Spaces:
Running
Running
File size: 2,846 Bytes
50f2856 4de8125 50f2856 4de8125 50f2856 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 | """Build the fixed task splits in data/tasks/.
python -m scott_wu_math.scripts.build_tasks # or: python scripts/build_tasks.py from the env dir
test: 50 deals per level (200), seed 2026
train: 1,000 deals per level (4,000), seed 7, none sharing an id with test
video: the two hands from the video, 2025 from eight cards and 6,843 from nine
"""
from __future__ import annotations
import json
import random
import sys
from pathlib import Path
HERE = Path(__file__).resolve().parents[1]
sys.path.insert(0, str(HERE.parent))
from scott_wu_math.game.cards import Card # noqa: E402
from scott_wu_math.game.expr import parse, steps, to_text # noqa: E402
from scott_wu_math.game.tasks import LEVELS, make_task # noqa: E402
OUT = HERE / "data" / "tasks"
# As played in the video (the suits are not all visible, so they are filled in).
VIDEO = [
{
"id": "video-2025",
"target": 2025,
"cards": [(13, "S"), (12, "H"), (11, "D"), (10, "C"), (7, "S"), (9, "H"), (5, "D"), (5, "C")],
"solution": "((13 - 12) + (11 - 10) + 7) * 9 * 5 * 5",
},
{
"id": "video-6843",
"target": 6843,
"cards": [(6, "S"), (4, "H"), (12, "D"), (10, "C"), (10, "S"), (8, "H"), (11, "D"), (7, "C"), (3, "S")],
"solution": "(((12 - 10) * (6 + 4) * 10 + 8) * 11 - 7) * 3",
},
]
def build(seed: int, per_level: int, prefix: str, skip: set[str]) -> list[dict]:
rng = random.Random(seed)
tasks: list[dict] = []
for level in LEVELS:
made = 0
while made < per_level:
task = make_task(rng, level=level)
key = task["id"]
if key in skip:
continue
skip.add(key)
task["id"] = f"{prefix}-{level}-{made:04d}"
task["key"] = key
tasks.append(task)
made += 1
return tasks
def write(name: str, tasks: list[dict]) -> None:
path = OUT / f"{name}.jsonl"
path.write_text("".join(json.dumps(t, ensure_ascii=False) + "\n" for t in tasks))
print(f"{path.relative_to(HERE)}: {len(tasks)} tasks")
def main() -> None:
OUT.mkdir(parents=True, exist_ok=True)
seen: set[str] = set()
test = build(2026, 50, "test", seen)
train = build(7, 1000, "train", seen)
video = []
for v in VIDEO:
cards = [Card(r, s) for r, s in v["cards"]]
tree = parse(v["solution"])
video.append(
{
"id": v["id"],
"level": "year" if len(cards) == 8 else "four-digit",
"target": v["target"],
"cards": [c.to_dict() for c in cards],
"solution": to_text(tree),
"solution_steps": steps(tree),
}
)
write("test", test)
write("train", train)
write("video", video)
if __name__ == "__main__":
main()
|