topsis的python代码

以下是使用Python实现的TOPSIS算法的代码示例：

import numpy as np

def topsis(data, weights, impacts):
    # 步骤1：归一化矩阵
    normalized_data = data / np.sqrt(np.sum(data**2, axis=0))
    
    # 步骤2：加权归一化矩阵
    weighted_normalized_data = normalized_data * weights
    
    # 步骤3：计算正理想解和负理想解
    positive_ideal = np.max(weighted_normalized_data, axis=1)
    negative_ideal = np.min(weighted_normalized_data, axis=1)
    
    # 步骤4：计算正负理想解之间的距离
    positive_distances = np.sqrt(np.sum((weighted_normalized_data - positive_ideal)**2, axis=1))
    negative_distances = np.sqrt(np.sum((weighted_normalized_data - negative_ideal)**2, axis=1))
    
    # 步骤5：计算接近度指数
    performance_scores = negative_distances / (positive_distances + negative_distances)
    
    # 根据正负影响进行调整
    for i in range(len(impacts)):
        if impacts[i] == '-':
            performance_scores[i] = 1 / (performance_scores[i] + 0.000001)  # 避免除零错误
        elif impacts[i] == '+':
            performance_scores[i] = performance_scores[i]
    
    return performance_scores

# 示例数据
data = np.array([[250, 16, 12],
                 [200, 12, 8],
                 [300, 20, 16],
                 [275, 18, 14]])
weights = np.array([0.4, 0.3, 0.3])
impacts = ['-','+','+']

# 使用TOPSIS算法计算指标的绩效得分
scores = topsis(data, weights, impacts)
print(scores)

请注意，上述代码中的示例数据仅为演示目的，您可以根据您自己的数据进行相应的调整。