mirror of
https://github.com/open-thought/reasoning-gym.git
synced 2026-04-19 12:58:07 +00:00
5d7fbac0ad
* math prompt improvements * ignore brackets in complex_arithmetic results * improve additional instruction in prompt of polynomial_equations * more strict tests for score_answer in polynomial_equations * simplify special reward handling * fix test_intermediate_integration * fix sokoban dataset * add common dataset score_answer consistency test
92 lines
3.1 KiB
Python
92 lines
3.1 KiB
Python
import json
|
|
|
|
import pytest
|
|
|
|
from reasoning_gym.algorithmic.game_of_life import GameOfLifeConfig, GameOfLifeDataset
|
|
|
|
|
|
def test_game_of_life_config_validation():
|
|
"""Test that invalid configs raise appropriate errors"""
|
|
with pytest.raises(AssertionError):
|
|
config = GameOfLifeConfig(grid_size_x=2) # Too small
|
|
config.validate()
|
|
|
|
with pytest.raises(AssertionError):
|
|
config = GameOfLifeConfig(grid_size_y=1000) # Too large
|
|
config.validate()
|
|
|
|
with pytest.raises(AssertionError):
|
|
config = GameOfLifeConfig(grid_size_x=5, grid_size_y=5, filled_cells=26) # Too many cells
|
|
config.validate()
|
|
|
|
|
|
def test_game_of_life_deterministic():
|
|
"""Test that dataset generates same items with same seed"""
|
|
config = GameOfLifeConfig(seed=42, size=10)
|
|
dataset1 = GameOfLifeDataset(config)
|
|
dataset2 = GameOfLifeDataset(config)
|
|
|
|
for i in range(len(dataset1)):
|
|
assert dataset1[i] == dataset2[i]
|
|
|
|
|
|
def test_game_of_life_basic_properties():
|
|
"""Test basic properties and solution of generated items"""
|
|
config = GameOfLifeConfig(seed=42, size=10, grid_size_x=20, grid_size_y=20, filled_cells=200, simulation_steps=1)
|
|
dataset = GameOfLifeDataset(config)
|
|
|
|
for item in dataset:
|
|
assert isinstance(item, dict)
|
|
assert "question" in item
|
|
assert "answer" in item
|
|
assert "metadata" in item
|
|
|
|
# Check metadata contains required fields
|
|
assert "grid_size_x" in item["metadata"]
|
|
assert "grid_size_y" in item["metadata"]
|
|
assert "filled_cells" in item["metadata"]
|
|
assert "simulation_steps" in item["metadata"]
|
|
|
|
# Test the scoring
|
|
assert dataset.score_answer(answer=item["answer"], entry=item) == 1.0
|
|
assert dataset.score_answer(answer=None, entry=item) == 0.0
|
|
assert dataset.score_answer(answer="invalid json", entry=item) == 0.0
|
|
|
|
config = GameOfLifeConfig(seed=43, size=1, grid_size_x=3, grid_size_y=3, filled_cells=1, simulation_steps=1)
|
|
dataset = GameOfLifeDataset(config)
|
|
|
|
for item in dataset:
|
|
assert isinstance(item, dict)
|
|
assert "question" in item
|
|
assert "answer" in item
|
|
assert "metadata" in item
|
|
|
|
ja = json.loads(item["answer"])
|
|
ja[0][0] = 1
|
|
ja[0][1] = 1
|
|
ja[0][2] = 1
|
|
jas = json.dumps(ja)
|
|
|
|
# Test the scoring
|
|
assert 0.1 < dataset.score_answer(answer=jas, entry=item) < 1.0
|
|
|
|
|
|
def test_game_of_life_iteration():
|
|
"""Test that iteration respects dataset size"""
|
|
config = GameOfLifeConfig(size=5, seed=42) # Small size for testing
|
|
dataset = GameOfLifeDataset(config)
|
|
|
|
# Test manual iteration
|
|
items = []
|
|
for item in dataset:
|
|
items.append(item)
|
|
assert len(items) == config.size, "Iterator should yield exactly size items"
|
|
|
|
# Test list conversion
|
|
items = list(dataset)
|
|
assert len(items) == config.size, "Iterator should yield exactly size items"
|
|
|
|
# Test multiple iterations
|
|
first_items = list(dataset)
|
|
second_items = list(dataset)
|
|
assert first_items == second_items, "Multiple iterations should yield same items"
|