python - 向后消除：“numpy.ndarray”对象没有属性“列”

Question

我正在运行一个用于向后消除的 while 循环，它将删除/迭代删除具有最大值的属性（建议：使用 p 值 < 0.05 的 while 循环）

thefile=input("Enter filename:")
separator=input("Enter symbol of separator (comma, semicolon etc):")
targetvar=input("Enter name of target variable for prediction "
                "(note: this is case sensitive): ") # MEDV for Boston.csv

# Test with boston.csv and other CSVs

# Load Dataset
dataframe=pd.read_csv(thefile,sep=separator)

# Show first few data points of the dataset
print("This is the preview of the loaded data:")
df = pd.read_csv(thefile,sep=separator)
X = dataframe.loc[:,dataframe.columns != targetvar].values
y = dataframe.iloc[:,targetvarloc].values
df.head()

#Backward Elimination
cols = list(X.columns)
pmax = 1
while (len(cols)>0):
    p= []
    X_1 = X[cols]
    X_1 = sm.add_constant(X_1)
    model = sm.OLS(y,X_1).fit()
    p = pd.Series(model.pvalues.values[1:],index = cols)      
    pmax = max(p)
    feature_with_p_max = p.idxmax()
    if(pmax>0.05):
        cols.remove(feature_with_p_max)
    else:
        break
selected_features_BE = cols
print(selected_features_BE)

Traceback (most recent call last)
<ipython-input-96-3f41adb6e314> in <module>()
----> 1 cols = list(X.columns)
      2 pmax = 1
      3 while (len(cols)>0):
      4     p= []
      5     X_1 = X[cols]

AttributeError: 'numpy.ndarray' object has no attribute 'columns'

标签： pythonpandasnumpydataframewhile-loop