#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ 熵减评估脚本 功能：评估决策的熵减效果，计算熵值 用途：当AI Agent需要评估决策的熵减效果时调用 参数：当前状态、目标状态、决策方案 输出：熵值计算、熵减效果评估、建议优化方案 """ import json import argparse from typing import Dict, List, Any from dataclasses import dataclass import math @dataclass class State: """状态定义""" name: str factors: Dict[str, float] # 因素名称 -> 因素值 (0-1, 0最有序, 1最混乱) structure_score: float # 结构化程度 (0-1, 0无结构, 1完全结构化) @dataclass class EntropyResult: """熵减评估结果""" current_entropy: float target_entropy: float entropy_reduction: float reduction_percentage: float factor_analysis: Dict[str, Any] recommendations: List[str] class EntropyAnalyzer: """熵减分析器""" def __init__(self): self.default_factors = { 'information_clarity': 0.7, # 信息清晰度 'process_order': 0.6, # 流程有序性 'decision_consistency': 0.5, # 决策一致性 'resource_allocation': 0.6, # 资源分配有序性 'system_stability': 0.7, # 系统稳定性 } def calculate_entropy(self, state: State) -> float: """ 计算状态的熵值 Args: state: 状态定义 Returns: float: 熵值 (0-1, 0最有序, 1最混乱) """ # 计算各因素的加权熵 factor_entropy = sum(state.factors.values()) / len(state.factors) # 结构化程度的反比作为结构熵 structure_entropy = 1.0 - state.structure_score # 综合熵值 total_entropy = 0.6 * factor_entropy + 0.4 * structure_entropy return min(max(total_entropy, 0.0), 1.0) def analyze(self, current_state: State, target_state: State, decision_description: str = "") -> EntropyResult: """ 分析熵减效果 Args: current_state: 当前状态 target_state: 目标状态 decision_description: 决策描述 Returns: EntropyResult: 熵减评估结果 """ # 计算当前熵 current_entropy = self.calculate_entropy(current_state) # 计算目标熵 target_entropy = self.calculate_entropy(target_state) # 计算熵减 entropy_reduction = current_entropy - target_entropy # 计算熵减百分比 reduction_percentage = (entropy_reduction / current_entropy * 100) if current_entropy > 0 else 0 # 因素分析 factor_analysis = self._analyze_factors(current_state, target_state) # 生成建议 recommendations = self._generate_recommendations(entropy_reduction, factor_analysis) return EntropyResult( current_entropy=current_entropy, target_entropy=target_entropy, entropy_reduction=entropy_reduction, reduction_percentage=reduction_percentage, factor_analysis=factor_analysis, recommendations=recommendations ) def _analyze_factors(self, current: State, target: State) -> Dict[str, Any]: """分析各因素的变化""" analysis = {} for factor_name in current.factors: current_value = current.factors.get(factor_name, 0.5) target_value = target.factors.get(factor_name, 0.5) improvement = current_value - target_value analysis[factor_name] = { 'current': current_value, 'target': target_value, 'improvement': improvement, 'improvement_percentage': (improvement / current_value * 100) if current_value > 0 else 0, 'impact': 'high' if improvement > 0.3 else 'medium' if improvement > 0.1 else 'low' } return analysis def _generate_recommendations(self, entropy_reduction: float, factor_analysis: Dict[str, Any]) -> List[str]: """生成优化建议""" recommendations = [] # 总体评估 if entropy_reduction > 0.5: recommendations.append("✓ 该决策能够实现显著的熵减，建议执行") elif entropy_reduction > 0.2: recommendations.append("✓ 该决策能够实现一定的熵减，建议进一步优化后执行") else: recommendations.append("✗ 该决策的熵减效果不明显，建议重新考虑") # 因素优化建议 high_impact_factors = [ name for name, data in factor_analysis.items() if data['impact'] == 'high' and data['improvement'] > 0 ] if high_impact_factors: recommendations.append(f"重点关注以下高影响因素: {', '.join(high_impact_factors)}") # 一般性建议 recommendations.extend([ "每个决策都应该问：这个选择是增加熵还是减少熵？", "优先选择能够建立最大秩序的方案", "通过结构优化、信息压缩、能量输入实现熵减", "主动对抗系统的熵增趋势" ]) return recommendations def create_state_from_dict(state_dict: Dict[str, Any]) -> State: """从字典创建状态""" return State( name=state_dict.get('name', 'unknown'), factors=state_dict.get('factors', {}), structure_score=state_dict.get('structure_score', 0.5) ) def main(): """主函数""" parser = argparse.ArgumentParser(description='熵减分析工具') parser.add_argument('--current-state', type=str, required=True, help='当前状态 (JSON格式)') parser.add_argument('--target-state', type=str, required=True, help='目标状态 (JSON格式)') parser.add_argument('--decision', type=str, default='', help='决策描述') parser.add_argument('--output', type=str, default='json', choices=['json', 'text'], help='输出格式') args = parser.parse_args() # 创建状态 current_state = create_state_from_dict(json.loads(args.current_state)) target_state = create_state_from_dict(json.loads(args.target_state)) # 创建分析器 analyzer = EntropyAnalyzer() # 执行分析 result = analyzer.analyze(current_state, target_state, args.decision) # 输出结果 if args.output == 'json': print(json.dumps({ 'current_entropy': result.current_entropy, 'target_entropy': result.target_entropy, 'entropy_reduction': result.entropy_reduction, 'reduction_percentage': f"{result.reduction_percentage:.2f}%", 'factor_analysis': result.factor_analysis, 'recommendations': result.recommendations }, ensure_ascii=False, indent=2)) else: print(f"当前熵值: {result.current_entropy:.3f}") print(f"目标熵值: {result.target_entropy:.3f}") print(f"熵减效果: {result.entropy_reduction:.3f} ({result.reduction_percentage:.2f}%)\n") print("因素分析:") for factor_name, data in result.factor_analysis.items(): print(f" {factor_name}:") print(f" 当前: {data['current']:.3f} -> 目标: {data['target']:.3f}") print(f" 改善: {data['improvement']:.3f} ({data['improvement_percentage']:.2f}%)") print(f" 影响: {data['impact']}") print("\n建议:") for i, recommendation in enumerate(result.recommendations, 1): print(f" {i}. {recommendation}") if __name__ == '__main__': main()