如何克服“'numpy.ndarray'对象不可调用”错误？

Question

我使用以下链接中的代码查看了使用 PCA 和自动编码器的异常检测：Machine learning for anomaly detection and condition monitoring 并且我尝试运行代码部分以使用带有 Mahalanobis 距离的 PCA，但是，如果我运行代码，我总是得到异常消息，事实证明问题出在协方差矩阵函数部分，其中出现错误'numpy.ndarray' object is not callable。我尝试创建新变量，将数据框更改为 NumPy，但没有任何效果导致此错误的原因是什么？

代码：

def cov_matrix(data, verbose=False):
    # data = pd.DataFrame(data).to_numpy()
    print('calculating the covaraince matrix')
    covariance_matrix = np.cov(data, rowvar=False)
    print('Done the covaraince matrix')
    if is_pos_def(covariance_matrix):
        inv_covariance_matrix = np.linalg.inv(covariance_matrix)
        if is_pos_def(inv_covariance_matrix):
            return covariance_matrix, inv_covariance_matrix
        else:
            print("Error: Inverse of Covariance Matrix is not positive definite!")
    else:
        print("Error: Covariance Matrix is not positive definite!")
        
def MahalanobisDist(inv_cov_matrix, mean_distr, data, verbose=False):
    inv_covariance_matrix = inv_cov_matrix
    vars_mean = mean_distr
    diff = data - vars_mean
    md = []
    for i in range(len(diff)):
        md.append(np.sqrt(diff[i].dot(inv_covariance_matrix).dot(diff[i])))
    return md

def MD_detectOutliers(dist, extreme=False, verbose=False):
    k = 3. if extreme else 2.
    threshold = np.mean(dist) * k
    outliers = []
    for i in range(len(dist)):
        if dist[i] >= threshold:
            outliers.append(i)  # index of the outlier
    return np.array(outliers)

def MD_threshold(dist, extreme=False, verbose=False):
    k = 3. if extreme else 2.
    threshold = np.mean(dist) * k
    return threshold

    #### Main code:
    # Inputting the training and test dataframes:
    data_train = np.array(principalDf_C0.values)
    data_test_C1 = np.array(principalDf_C1.values)
    data_test_C2 = np.array(principalDf_C2.values)
    data_test_C3 = np.array(principalDf_C4.values)
    data_test_C4 = np.array(principalDf_C5.values)
    
    print('Training Dataframe: ', data_train[:,])
    print('Test1 Dataframe: ', data_test_C1)
    print('Test2 Dataframe: ', data_test_C2)
    print('Test3 Dataframe: ', data_test_C3)
    print('Test4 Dataframe: ', data_test_C4)
    
    data_train_df = pd.DataFrame(principalDf_C0.values)
    data_test_df_C1 =  pd.DataFrame(principalDf_C1.values)
    data_test_df_C2 =  pd.DataFrame(principalDf_C2.values)
    data_test_df_C3 =  pd.DataFrame(principalDf_C4.values)
    data_test_df_C4 =  pd.DataFrame(principalDf_C5.values)
    
    # Calculating the covariance matrix:
    cov_matrix, inv_cov_matrix = cov_matrix(data=data_train)
    
    # Calculating the mean value for the input variables:
    mean_distr = data_train_df.mean(axis=0)
    
    # Calculating the Mahalanobis distance and threshold value to flag datapoints as an anomaly:
    dist_test_C1 = MahalanobisDist(inv_cov_matrix, mean_distr, data_test_df_C1, verbose=True)
    dist_test_C2 = MahalanobisDist(inv_cov_matrix, mean_distr, data_test_df_C2, verbose=True)
    dist_test_C3 = MahalanobisDist(inv_cov_matrix, mean_distr, data_test_df_C3, verbose=True)
    dist_test_C4 = MahalanobisDist(inv_cov_matrix, mean_distr, data_test_df_C4, verbose=True)
    dist_train = MahalanobisDist(inv_cov_matrix, mean_distr, data_train_df, verbose=True)
    threshold = MD_threshold(dist_train, extreme = True)

    # Distribution of Threshold value for flagging an anomaly:
    plt.figure()
    sns.distplot(np.square(dist_train),bins = 10, kde= False)
    # plt.xlim([0.0,15])
    plt.show()
    
    plt.figure()
    sns.distplot(dist_train, bins = 10, kde= True, color = 'green');
    # plt.xlim([0.0,5])
    plt.xlabel('Mahalanobis dist')
    plt.show()
    
    anomaly_train = pd.DataFrame(index=data_train_df.index)
    anomaly_train['Mob dist']= dist_train
    anomaly_train['Thresh'] = threshold
    # If Mob dist above threshold: Flag as anomaly
    anomaly_train['Anomaly'] = anomaly_train['Mob dist'] > anomaly_train['Thresh']
    anomaly_train.index = X_train_PCA.index
    
    anomaly_C1 = pd.DataFrame(index=data_test_df_C1.index)
    anomaly_C1['Mob dist']= dist_test_C1
    anomaly_C1['Thresh'] = threshold
    # If Mob dist above threshold: Flag as anomaly
    anomaly_C1['Anomaly'] = anomaly_C1['Mob dist'] > anomaly_C1['Thresh']
    anomaly_C1.index = data_test_df_C1.index
    anomaly_C1.head()
    
    anomaly_C2 = pd.DataFrame(index=data_test_df_C2.index)
    anomaly_C2['Mob dist']= dist_test_C2
    anomaly_C2['Thresh'] = threshold
    # If Mob dist above threshold: Flag as anomaly
    anomaly_C2['Anomaly'] = anomaly_C2['Mob dist'] > anomaly_C2['Thresh']
    anomaly_C2.index = data_test_df_C2.index
    anomaly_C2.head()
    
    anomaly_C3 = pd.DataFrame(index=data_test_df_C3.index)
    anomaly_C3['Mob dist']= dist_test_C3
    anomaly_C3['Thresh'] = threshold
    # If Mob dist above threshold: Flag as anomaly
    anomaly_C3['Anomaly'] = anomaly_C3['Mob dist'] > anomaly_C3['Thresh']
    anomaly_C3.index = data_test_df_C3.index
    anomaly_C3.head()
    
    anomaly_C4 = pd.DataFrame(index=data_test_df_C4.index)
    anomaly_C4['Mob dist']= dist_test_C4
    anomaly_C4['Thresh'] = threshold
    # If Mob dist above threshold: Flag as anomaly
    anomaly_C4['Anomaly'] = anomaly_C4['Mob dist'] > anomaly_C4['Thresh']
    anomaly_C4.index = data_test_df_C4.index
    anomaly_C4.head()

    final_scored = pd.concat([anomaly_train, anomaly_C1, anomaly_C2, anomaly_C3, anomaly_C4])
    print(final_scored)
except Exception:
    print('Cannot implement Anomaly detection using Mahalanobis distance metric')
    pass

Answer 1

根据您的评论，您在 var cov_matrix 和函数 cov_matrix() 之间存在命名空间冲突

将该行更改为例如

matrix, inv_matrix = cov_matrix(data=data_train)

并相应地更新您的代码，或重命名cov_matrix()。一个好的约定是返回事物的函数的名称中应该包含动词，例如generate_cov_matrix() 或 calculate_cov_matrix().*

（是的，代码应该运行一次，因为 AFAICS 之后您不会再调用 cov_matrix()，但我猜您正在使用持久解释器会话并再次评估代码一次 cov_matrix()已被覆盖。）

*此约定假定函数存在副作用，并异常返回。当然，如果你是在写函数式的，并且有副作用是例外而不是规则，你可能想要颠倒它，或者完全遵循另一个约定。

Answer 2

我的猜测是您遇到了一个问题，即您有一个名为 cov_matrix 的变量和一个名为 cov_matrix 的函数。在某些时候，我认为你用变量覆盖了函数，这是一个numpy.ndarray。稍后你尝试调用函数cov_matrix()，但对象实际上是变量，即numpy数组。