获取LinAlgError:奇异矩阵错误 [英] Getting LinAlgError: Singular matrix Error

问题描述

我正在使用以下函数来计算p值，以建立拟合逻辑回归模型.但是我得到了LinAlgError:奇异矩阵错误.

I'm using the below function to calculate p-value to build fit logistic regression model. But I get LinAlgError: Singular matrix error.

from sklearn import linear_model
import scipy.stats as stat

class LogisticRegression_with_p_values:
    def __init__(self, *args, **kwargs):
      self.model = linear_model.LogisticRegression(*args, **kwargs)
    def fit(self, X, y):
      self.model.fit(X, y)
      denom = (2.0 * (1.0 + np.cosh(self.model.decision_function(X))))
      denom = np.tile(denom, (X.shape[1],1)).T
      F_ij = np.dot((X/denom).T,X)
      Cramer_Rao = np.linalg.inv(F_ij)
      sigma_estimates = np.sqrt(np.diagonal(Cramer_Rao))
      z_scores = self.model.coef_[0] / sigma_estimates
      p_values = [stat.norm.sf(abs(x)) * 2 for x in z_scores]
      self.coef_ = self.model.coef_
      self.intercept_ = self.model.intercept_
      self.p_values = p_values
reg = LogisticRegression_with_p_values()
reg.fit(inputs_train, loan_data_targets_train)

推荐答案

错误:reg = LogisticRegression_with_p_values()LinAlgError:奇异矩阵P值函数抛出错误后拟合模型:LinAlgError:奇异矩阵

Error : reg = LogisticRegression_with_p_values() LinAlgError: Singular matrix Fitting the Model after the P-value function throws error: LinAlgError: Singular matrix

步骤1:运行以下代码，并观察绿线中的所有缺失值.

Step 1: run the below code and observe any missing values in the green line.

corr = inputs_train.corr() 

kot = corr[corr>=.9] 

plt.figure(figsize=(18,10)) 

sns.heatmap(kot, cmap="Greens")

我正在研究Lending Club分析并遇到此错误，因此使用了上述热图，并在绿线中找到了一个缺失值，因此为了进一步调查，我运行了以下代码，以检查输出是否为"nan" inputs_train ['term:36'].corr(inputs_train ['term:36'])

I am working on Lending Club analysis and encountered this error so used the above-mentioned heatmap and found a missing value in the green line so to further investigate I ran the below code to check if the output is 'nan' inputs_train['term:36'].corr(inputs_train['term:36'])

O/P ::南

下一步:

当我们创建变量'term_int'时，我们没有将其从字符串转换为数字.要验证此检查输出，如果type(df_inputs_prepr_train ['term_int'] [0])是否提供'STR'

When we create the variable 'term_int', we have not converted it to numerical from string. To verify this check output if type(df_inputs_prepr_train['term_int'][0]) gives 'STR'

所以当我们写的时候:

df_inputs_prepr_train[df_inputs_prepr_train['term_int']==36]['term_int']

它显示输出为零行，因为'term_int'仍然是str 36，而不是数字36.

it shows the output as zero rows since 'term_int' is still an str 36 and not a numerical 36.

所以当我们使用代码:::

so when we use the code ::

 df_inputs_prepr_train['term:36']=np.where((df_inputs_prepr_train['term_int']==36),1,0)

  it basically stores only zeroes as output. 

要采取的措施::

`df_inputs_prepr_train['term_int']=pd.to_numeric(df_inputs_prepr_train['term_int'])`

交叉验证:如果再次运行热图，绿线中将看不到任何缺失的值

Cross-verify: if you run heatmap again you won't see any missing values in the green line

这篇关于获取LinAlgError:奇异矩阵错误的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！