After LinearRegression Model, How to Pair the Coefficients with the corresponding Features?-CodePudding

I built my first LinearRegression model (ElasticNet), predicting house SalePrice.

I would like to find out the features that have strong correlations (both negative and positive correlations) with the SalePrice

In the screenshot, I listed out all the coefficient and feature names. What code can I use to pair these two values so I can see each feature's coefficient value?

I am very new to coding and data analytics. Thank you in advance!

My model:

grid_model = GridSearchCV(estimator = base_elastic_model,
                     param_grid = param_grid,
                     scoring = 'neg_mean_squared_error',
                     cv=5,
                     verbose=1)
grid_model.fit(scaled_X_train,y_train)

I got the list of coefficient:

grid_model.fit(scaled_X_train,y_train)

I got the list of features whose coefficent with the SalePrice is not 0

df.columns[coef[coef == 0].index]

How can i print a dataframe with Coefficient and Feature Name listed matching each other?

CodePudding user response：

Try this:

pd.DataFrame(X_train.columns, grid_model.best_estimator_.coef_)

It will give output like this:

-0.003801   feature0
-0.033107   feature1
0.053203    feature2
-0.645900   feature3
-7.474264   feature4
-0.571417   feature5
0.007333    feature6
0.184133    feature7
0.091905    feature8
0.002021    feature9

CodePudding user response：

-0.003801 feature0 -0.033107 feature1 0.053203 feature2 -0.645900 feature3 -7.474264 feature4 -0.571417 feature5 0.007333 feature6 0.184133 feature7 0.091905 feature8 0.002021 feature9