File size: 3,333 Bytes
ce7bd54 | 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 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 | """
Number Game Reward Function
评分规则：
- 选择正确的数字: +1.0
- 选择错误的数字: 0.0
输入格式:
reward_input = {
"response": "1", # 模型输出的答案 (0/1/2)
"response_length": 10, # 响应长度（token数）
"ground_truth": "1" # 正确答案 (0/1/2)
}
输出格式:
{
"overall": 1.0, # 总分（必需字段）
"accuracy": 1.0 # 准确率（可选，用于监控）
}
"""
import re
from typing import Any
# Metadata - EasyR1框架要求
REWARD_NAME = "number_game"
REWARD_TYPE = "batch" # 批量处理模式
def extract_answer(response: str) -> str:
"""
从模型响应中提取答案索引
Args:
response: 模型的原始响应
Returns:
"0", "1", "2" 或 ""（提取失败）
"""
# 情况1: 响应本身就是单个数字
response = response.strip()
if response in ["0", "1", "2"]:
return response
# 情况2: 响应包含多余文字，提取第一个出现的0/1/2
match = re.search(r"[012]", response)
if match:
return match.group(0)
# 提取失败
return ""
def compute_score(reward_inputs: list[dict[str, Any]]) -> list[dict[str, float]]:
"""
计算一批样本的得分
Args:
reward_inputs: 包含多个样本的列表，每个样本包含:
- response: 模型的响应
- response_length: 响应长度
- ground_truth: 正确答案
Returns:
每个样本的得分字典列表，包含:
- overall: 总分（1.0表示正确，0.0表示错误）
- accuracy: 准确率（同overall，用于监控）
"""
scores = []
for reward_input in reward_inputs:
response = reward_input.get("response", "")
ground_truth = reward_input.get("ground_truth", "")
# 提取答案
predicted = extract_answer(response)
# 计算得分
if predicted == ground_truth:
score = 1.0
else:
score = 0.0
# 返回格式：必须包含overall字段
scores.append({"overall": score, "accuracy": score})
return scores
# 测试用例
if __name__ == "__main__":
test_cases = [
# 完美匹配
{"response": "0", "response_length": 1, "ground_truth": "0"},
{"response": "1", "response_length": 1, "ground_truth": "1"},
{"response": "2", "response_length": 1, "ground_truth": "2"},
# 响应包含额外文字
{"response": "The answer is 1", "response_length": 15, "ground_truth": "1"},
{"response": "I choose option 2", "response_length": 18, "ground_truth": "2"},
# 错误答案
{"response": "0", "response_length": 1, "ground_truth": "1"},
{"response": "2", "response_length": 1, "ground_truth": "0"},
# 提取失败
{"response": "I don't know", "response_length": 12, "ground_truth": "1"},
{"response": "", "response_length": 0, "ground_truth": "2"},
]
scores = compute_score(test_cases)
print("Reward Function Test Results:")
print("=" * 60)
for i, (test, score) in enumerate(zip(test_cases, scores), 1):
print(f"{i}. Response: {test['response']!r}")
print(f" Ground Truth: {test['ground_truth']!r}")
print(f" Score: {score}")
print()
|