python 代码实现了一个结合数据包络分析（DEA）和粒子群优化（PSO）算法的模型，主要用于寻找一组最优权重，以评估和优化某种与企业生产效率相关的指标

import numpy as np
import pandas as pd
import pickle
import re
from scipy.optimize import minimize
import random
import matplotlib.pyplot as plt
import scipy.stats as stats
from concurrent.futures import ThreadPoolExecutor


# 加载数据，添加文件存在性验证
def load_data():
    try:
        with open('fited_weight_2018.pkl', 'rb') as f:
            fited_weight = pickle.load(f)
        with open('BothExit_symbol_2018.pkl', 'rb') as f:
            both_exit_symbol = pickle.load(f)
        df_profit = pd.read_csv('./PT_LCMAINFIN_new.csv', header=0)
        df_profit = df_profit.dropna(subset=['IndustryCode'])
        df_profit['IndustryCode'] = df_profit['IndustryCode'].astype(float).astype(int)
        df_profit['EndDate'] = df_profit['EndDate'].str.slice(0, 4)
        df_profit = df_profit[["Symbol", "EndDate", "IndustryCode", "TotalAssets", "OperatingCost", "OperatingEvenue"]]
        df_profit['EndDate'] = df_profit['EndDate'].astype(float).astype(int)
        df_profit_2018 = df_profit[df_profit['EndDate'] == 2018]
        df = pd.read_csv('D:\\碳文献\\碳价\\汇总数据\\上市公司行业分类.csv', header=0)
        df_employ = pd.read_csv('./13.csv', header=0)
        df_employ = df_employ.dropna(subset=['Employment'])
        return fited_weight, both_exit_symbol, df_profit_2018, df, df_employ
    except FileNotFoundError as e:
        print(f"文件未找到: {
     e}")
        raise


# 初始化粒子
def enable(particles):
    for particle in particles:
        particle['weight'] = np.array(rand_weight())
        particle['pbest'] = particle['weight']
        particle['v'] = np.array([0.0001] * 4)
    return particles


# 生成随机权重
def rand_weight():
    weights = np.random.uniform(0