init-commit

internbootcamp/bootcamp/kor_operation_unicode221e/kor_operation_unicode221e.py

@@ -0,0 +1,292 @@
+"""# 谜题训练场开发任务
+
+## 任务概述
+你是一位资深程序员，我需要你帮我实现一个特定谜题的训练场环境类。这个类继承自`Basebootcamp`，用于生成谜题实例并验证解答。
+
+## 背景说明
+我正在开发一系列谜题训练场，每个训练场对应一个特定类型的谜题。训练场类命名为`{PuzzleName}bootcamp`，其中`PuzzleName`是谜题的名称。
+
+每个训练场类主要提供两个核心功能：
+1. 生成该谜题类型的问题实例
+2. 验证用户对问题的回答是否正确
+
+## 技术接口规范
+
+### 类方法实现要求
+
+```python
+from bootcamp import Basebootcamp
+
+class {PuzzleName}bootcamp(Basebootcamp):
+    def __init__(self, **params):
+        \"\"\"
+        请你自定义params，以保存该puzzle相关的参数，例如网格大小等，参数配有默认值
+        \"\"\"
+        pass
+    
+    def case_generator(self):
+        \"\"\"
+        生成谜题实例，提示：为保证谜题有解，可以先生成结果再对结果处理得到谜题
+        返回：一个可JSON序列化的字典（避免包含set等无法通过json.dumps处理的数据结构）
+        \"\"\"
+        pass
+    
+    @staticmethod
+    def prompt_func(question_case) -> str:
+        \"\"\"
+        将case_generator生成的谜题实例转换为文本形式的问题，问题中包含问题背景、对谜题规则的介绍、具体要解决的谜题实例、期望最终答案的格式，
+        例如：你是xxxx，请你解答yyyy，规则如下：yyyy，最终答案放置在：zzzzz
+        注意：请参照提供的谜题描述进行复述，规则应当描述详细，包括任务背景、具体任务操作规则、对题目格式和答案格式的含义介绍等，
+
+        参数:
+            question_case: 由case_generator生成的谜题实例
+            
+        返回:
+            str: 格式化的问题字符串
+            
+        注意:
+            1. 需考虑问题的格式，以便后续能正确提取
+            2. 问题描述中应包含期望的答案格式说明，以便后续能正确提取，为了避免抽取时匹配出干扰项，请要求模型将答案放在特定标签（如双括号）内，例如[[your answer here]]
+        \"\"\"
+        pass
+    
+    @staticmethod
+    def extract_output(output):
+        \"\"\"
+        从LLM的回复中提取符合格式要求的答案，如有多个，请抽取最后一个，避免使用re.search等只抽取第一个结果的方式。
+        
+        参数:
+            output: LLM的完整输出（包含原始问题和回答）
+            
+        返回:
+            提取的答案，若未找到符合格式的答案则返回None
+        \"\"\"
+        pass
+    
+    @classmethod
+    def _verify_correction(cls, solution, identity):
+        \"\"\"
+        验证提取的答案是否正确，注意一个问题可以能有多个解，按照谜题规则进行检验，不要直接匹配可能的答案。
+        
+        参数:
+            solution: extract_output提取的答案
+            identity: case_generator生成的谜题实例
+            
+        返回:
+            bool: 答案是否正确
+        \"\"\"
+        pass
+```
+
+### 验证评分方法（基类已实现）
+
+```python
+@classmethod
+def verify_score(cls, model_output, identity:dict, format_score=0.1) -> float:
+    \"\"\"
+    验证输出结果并评分。
+    
+    参数:
+        model_output: 模型的完整输出
+        identity: 谜题实例（由case_generator生成）
+        format_score: 答案格式正确时的基础分数
+    
+    返回:
+        float: 评分结果（0-1之间）
+    \"\"\"
+    score = 0. 
+    try:
+        extract_solution = cls.extract_output(model_output)
+        if extract_solution is None:
+            return score
+        else:
+            score = format_score # 格式正确时的基础分数
+        if cls._verify_correction(extract_solution, identity):
+            score = 1.  # 答案完全正确时的满分
+    except Exception as e:
+        # 处理异常情况
+        pass
+    return score
+```
+
+### 使用示例
+
+```python
+# 初始化谜题训练场
+bootcamp = Puzzlebootcamp()
+
+# 生成谜题实例
+case = bootcamp.case_generator()
+
+# 将谜题转换为文本问题
+prompt = Puzzlebootcamp.prompt_func(case)
+
+# 获取LLM对问题的解答
+response = get_response(prompt, \"LLM\")
+
+# 从完整对话中提取答案
+extracted_output = Puzzlebootcamp.extract_output(prompt + response)
+
+# 验证答案并评分
+score = Puzzlebootcamp.verify_score(extracted_output, case)
+```
+
+## 你的任务
+请根据以下谜题描述（谜题描述可能不完整，请先结合你的知识澄清规则），实现一个完整的谜题训练场类：
+
+### 谜题描述
+a∞b=a^2+b^2.
+a and b are positive integers.Example questions are as follows:
+
+<example 0>
+Compute 2∞3.
+Please ensure the answer is a single number and wrap it in double square brackets, like this: [[your answer]].
+</example 0>
+
+<example 1>
+Compute 4∞5.
+Please ensure the answer is a single number and wrap it in double square brackets, like this: [[your answer]].
+</example 1>
+
+<example 2>
+Compute 1∞7.
+Please ensure the answer is a single number and wrap it in double square brackets, like this: [[your answer]].
+</example 2>
+
+<example 3>
+Compute 3∞6.
+Please ensure the answer is a single number and wrap it in double square brackets, like this: [[your answer]].
+</example 3>
+
+<example 4>
+Compute 5∞8.
+Please ensure the answer is a single number and wrap it in double square brackets, like this: [[your answer]].
+</example 4>
+
+<example 5>
+If X∞4=20, find X.
+The answer should only be given as a number.
+Please wrap the answer in double square brackets, like this: [[your answer]].
+</example 5>
+
+<example 6>
+If 3∞X=18, find X.
+The answer should only be given as a number.
+Please wrap the answer in double square brackets, like this: [[your answer]].
+</example 6>
+
+<example 7>
+If X∞5=41, find X.
+The answer should only be given as a number.
+Please wrap the answer in double square brackets, like this: [[your answer]].
+</example 7>
+
+<example 8>
+If X∞8=80, find X.
+The answer should only be given as a number.
+Please wrap the answer in double square brackets, like this: [[your answer]].
+</example 8>
+
+<example 9>
+If X∞6=45, find X.
+The answer should only be given as a number.
+Please wrap the answer in double square brackets, like this: [[your answer]].
+</example 9>
+
+
+请完成上述谜题的训练场环境类实现，包括所有必要的方法。
+"""
+
+from bootcamp import Basebootcamp
+import re
+import random
+from bootcamp import Basebootcamp
+
+class KorOperationUnicode221ebootcamp(Basebootcamp):
+    def __init__(self, max_num=10, solve_ratio=0.5):
+        self.max_num = max_num
+        self.solve_ratio = solve_ratio
+    
+    def case_generator(self):
+        """增强案例生成逻辑，保证数值有效性"""
+        if random.random() < self.solve_ratio:
+            var_type = random.choice(['x', 'y'])
+            if var_type == 'x':
+                while True:
+                    b = random.randint(1, self.max_num)
+                    x = random.randint(1, self.max_num)
+                    c = x**2 + b**2
+                    # 确保解在允许范围内
+                    valid_solutions = [
+                        i for i in range(1, self.max_num+1)
+                        if i**2 + b**2 == c
+                    ]
+                    if len(valid_solutions) == 1:
+                        return {"problem_type": "solve_x", "b": b, "c": c}
+            else:
+                while True:
+                    a = random.randint(1, self.max_num)
+                    y = random.randint(1, self.max_num)
+                    c = a**2 + y**2
+                    valid_solutions = [
+                        i for i in range(1, self.max_num+1)
+                        if a**2 + i**2 == c
+                    ]
+                    if len(valid_solutions) == 1:
+                        return {"problem_type": "solve_y", "a": a, "c": c}
+        else:
+            # 允许生成相同的a和b
+            a = random.randint(1, self.max_num)
+            b = random.randint(1, self.max_num)
+            return {"problem_type": "compute", "a": a, "b": b}
+    
+    @staticmethod
+    def prompt_func(question_case) -> str:
+        problem_templates = {
+            "compute": "计算 {a}∞{b} 的结果。答案格式示例：[[13]]",
+            "solve_x": "解方程 X∞{b} = {c}。答案格式示例：[[2]]",
+            "solve_y": "解方程 {a}∞Y = {c}。答案格式示例：[[3]]"
+        }
+        template = """请根据以下规则解决问题：
+- 运算符定义：a∞b = a² + b²
+- 所有变量均为正整数
+
+当前问题：
+{problem_statement}
+
+请将最终答案放在双括号内，如：[[答案]]"""
+        
+        return template.format(
+            problem_statement=problem_templates[question_case["problem_type"]].format(**question_case)
+        )
+    
+    @staticmethod
+    def extract_output(output):
+        """增强正则匹配模式"""
+        matches = re.findall(r'\[\[\s*(\d+)\s*\]\]', output)
+        return matches[-1] if matches else None
+    
+    @classmethod
+    def _verify_correction(cls, solution, identity):
+        """增强数值范围验证"""
+        try:
+            sol = int(solution)
+            if sol < 1:
+                return False
+        except (ValueError, TypeError):
+            return False
+        
+        if identity["problem_type"] == "compute":
+            a, b = identity["a"], identity["b"]
+            return sol == a**2 + b**2
+        
+        if identity["problem_type"] == "solve_x":
+            b_val = identity["b"]
+            c_val = identity["c"]
+            return sol**2 + b_val**2 == c_val and sol <= identity.get("max_num", 10)
+        
+        if identity["problem_type"] == "solve_y":
+            a_val = identity["a"]
+            return a_val**2 + sol**2 == identity["c"] and sol <= identity.get("max_num", 10)
+        
+        return False