InternBootcamp/internbootcamp/bootcamp/tapa/tapa.py
2025-05-23 15:27:15 +08:00

344 lines
13 KiB
Python
Executable file
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

"""# 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
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
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[answer] [/answer]内
\"\"\"
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
**Tapa Puzzle Rules**
**Objective**: Blacken cells on a grid to satisfy all clues while adhering to connectivity and area constraints.
1. **Clues**:
- Each white cell containing a clue has one or more numbers (e.g., \"3\", \"1 2\").
- Numbers represent the lengths of **orthogonally connected black cell groups** in the 8 surrounding cells (up, down, left, right, and diagonals).
- Multiple numbers (e.g., \"2 1\") indicate **separate groups**, each isolated by at least one white cell.
2. **Group Formation**:
- A \"group\" is a set of black cells connected orthogonally (horizontally/vertically), **not diagonally**.
- Example: A clue \"3\" requires three orthogonally connected black cells in its perimeter. A clue \"1 2\" requires one isolated black cell and a separate pair of orthogonally connected black cells.
3. **Global Constraints**:
- **Single Connected Region**: All black cells must form one connected area (diagonals allowed for connectivity).
- **No 2×2 Black Blocks**: No 2×2 square can be entirely black.
- **No Isolation**: White cells must not be fully enclosed by black cells (i.e., all white cells must be reachable from the grids edges).
**Key Notes**:
- Clue cells themselves remain **white**.
- Numbers in clues can appear in any order (e.g., \"1 2\" and \"2 1\" are equivalent).
- A clue cell with \"0\" means none of its 8 surrounding cells are black.
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
from collections import deque
import json
class Tapabootcamp(Basebootcamp):
def __init__(self, rows=5, cols=5):
super().__init__()
self.rows = rows
self.cols = cols
def case_generator(self):
# 示例解:中间为黑,其他为白
solution = [[False for _ in range(self.cols)] for _ in range(self.rows)]
center_r, center_c = self.rows//2, self.cols//2
solution[center_r][center_c] = True
clues = {}
for r in range(self.rows):
for c in range(self.cols):
if not solution[r][c]:
groups = self.get_clue_numbers(solution, r, c, self.rows, self.cols)
if groups != [0]:
clues[f"{r},{c}"] = groups
return {
'rows': self.rows,
'cols': self.cols,
'clues': clues
}
@staticmethod
def prompt_func(question_case) -> str:
rows = question_case['rows']
cols = question_case['cols']
clues = question_case['clues']
prompt = "You are a solver for Tapa puzzles. Your task is to blacken cells in a grid according to the given clues and rules.\n\n"
prompt += "**Rules**:\n"
prompt += "- Each clue is a white cell with numbers indicating the lengths of orthogonally connected black cell groups in the surrounding 8 cells.\n"
prompt += "- Multiple numbers indicate separate groups, each isolated by at least one white cell.\n"
prompt += "- All black cells must form a single connected region (diagonally allowed).\n"
prompt += "- No 2×2 area can be entirely black.\n"
prompt += "- White cells must not be enclosed by black cells; they must be reachable from the grid's edge.\n\n"
prompt += f"The puzzle grid is {rows}x{cols}. The clues are as follows:\n"
for r_c, numbers in clues.items():
row, col = map(int, r_c.split(','))
nums_str = ' '.join(map(str, numbers))
prompt += f"- Cell at row {row}, column {col}: {nums_str}\n"
prompt += "\nYour answer should be a 2D list where each element is True (black) or False (white), enclosed within [answer] tags. For example:\n[answer]\n[[False, True], [True, False]]\n[/answer]"
return prompt
@staticmethod
def extract_output(output):
pattern = r'\[answer\](.*?)\[/answer\]'
matches = re.findall(pattern, output, re.DOTALL)
if not matches:
return None
last_match = matches[-1].strip()
try:
solution = eval(last_match)
if isinstance(solution, list) and all(isinstance(row, list) for row in solution):
return solution
return None
except:
return None
@classmethod
def _verify_correction(cls, solution, identity):
rows = identity['rows']
cols = identity['cols']
clues = identity['clues']
if not isinstance(solution, list) or len(solution) != rows:
return False
for row in solution:
if not isinstance(row, list) or len(row) != cols:
return False
for r_c, numbers in clues.items():
row, col = map(int, r_c.split(','))
if solution[row][col]:
return False
computed = cls.get_clue_numbers(solution, row, col, rows, cols)
if sorted(numbers) != sorted(computed):
return False
black_cells = [(r, c) for r in range(rows) for c in range(cols) if solution[r][c]]
if black_cells:
if not cls.is_connected(black_cells, rows, cols):
return False
for r in range(rows - 1):
for c in range(cols - 1):
if solution[r][c] and solution[r][c+1] and solution[r+1][c] and solution[r+1][c+1]:
return False
white_cells = [(r, c) for r in range(rows) for c in range(cols) if not solution[r][c]]
if not white_cells:
return False
visited = set()
queue = deque()
for (r, c) in white_cells:
if r == 0 or r == rows-1 or c == 0 or c == cols-1:
if (r, c) not in visited:
queue.append((r, c))
visited.add((r, c))
while queue:
r, c = queue.popleft()
for dr, dc in [(-1,0), (1,0), (0,1), (0,-1)]:
nr, nc = r + dr, c + dc
if 0 <= nr < rows and 0 <= nc < cols:
if not solution[nr][nc] and (nr, nc) not in visited:
visited.add((nr, nc))
queue.append((nr, nc))
if any((r, c) not in visited for (r, c) in white_cells):
return False
return True
@staticmethod
def get_clue_numbers(solution, row, col, rows, cols):
if solution[row][col]:
return []
directions = [(-1, -1), (-1, 0), (-1, 1),
(0, -1), (0, 1),
(1, -1), (1, 0), (1, 1)]
adjacent = []
for dr, dc in directions:
r = row + dr
c = col + dc
if 0 <= r < rows and 0 <= c < cols:
adjacent.append((r, c))
black_cells = [(r, c) for (r, c) in adjacent if solution[r][c]]
if not black_cells:
return [0]
visited = set()
groups = []
for (r, c) in black_cells:
if (r, c) not in visited:
queue = deque([(r, c)])
visited.add((r, c))
size = 1
while queue:
x, y = queue.popleft()
for dx, dy in [(-1,0), (1,0), (0,1), (0,-1)]:
nx, ny = x + dx, y + dy
if (nx, ny) in black_cells and (nx, ny) not in visited:
visited.add((nx, ny))
queue.append((nx, ny))
size += 1
groups.append(size)
groups.sort()
return groups if groups else [0]
@classmethod
def is_connected(cls, cells, rows, cols):
if not cells:
return True
start = cells[0]
visited = set([start])
queue = deque([start])
while queue:
r, c = queue.popleft()
for dr in [-1, 0, 1]:
for dc in [-1, 0, 1]:
if dr == 0 and dc == 0:
continue
nr, nc = r + dr, c + dc
if (nr, nc) in cells and (nr, nc) not in visited:
visited.add((nr, nc))
queue.append((nr, nc))
return len(visited) == len(cells)