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

Question

Here is my program using sklearn.

 X = np.array([[1, 2, 4],[2, 3, 9]]).T    
 print(X)
 y = np.array([1, 4, 16])
 X_poly = PolynomialFeatures(degree=2).fit_transform(X)
 print(X_poly)
 model = LinearRegression(fit_intercept = False)
 model.fit(X_poly,y)
 print('Coefficients: \n', model.coef_)
 print('Others: \n', model.intercept_)
 print(X_poly.powers_)
 X_predict = np.array([[3,3]])
 print(model.predict(feats.transform(X_predict)))

I have these errors:

 ---> 17 print(X_poly.powers_)
 18 
 19 X_predict = np.array([[3,3]])

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

Any help please ?

Answer 1

Splitting X_poly into two lines should do the trick here:

X_poly_temp = PolynomialFeatures(degree=2)
X_poly = X_poly_temp.fit_transform(X)
print(X_poly_temp.powers_)

Answer 2

Try modifying the line

print(X_poly.powers_)

To

print(X_poly[0].powers_)

If the error disappears then loop through X_poly and print the value for each index of the array.

Answer 3

Split fitting into two lines, one for initializing other for fitting, also save the return value of the fit step so you can use it to train the LinearRegression model.

X_poly = PolynomialFeatures(degree=2)
X_poly_return = X_poly.fit_transform(X)
print(X_poly)
model = LinearRegression(fit_intercept = False)
model.fit(X_poly_return,y)
print('Coefficients: \n', model.coef_)
print('Others: \n', model.intercept_)
print(X_poly.powers_)
X_predict = np.array([[3,3]])