In [7]:
#Making predictions using the 'model' created above

predict = np.poly1d(model)
Ad_Expenditure = 51
predict(Ad_Expenditure)
Out[7]:
872.9716236112969