a bit crazy

With lots of sadness in mind… the time doesn stop, so I have to go for my works

I’m involved in a awesome course by fun-mooc.fr (in English of course) and the question was:

Use a DummyClassifier such that the resulting classifier will always predict the class ' >50K'. What is the accuracy score on the test set? Repeat the experiment by always predicting the class ' <=50K'.

and I did


from sklearn.dummy import DummyClassifier

dummymodel = DummyClassifier(strategy='prior')
_ = dummymodel.fit(data_train,target_train)


# for checking if any False is available (false referse to ' >50K')
for i in (dummymodel.predict(data_test) == ' <=50K'):
    if i == False:
        print("there is a false")
 
print("This model predicts only ' <=50K'")
print("Score based on train data is\t{}".format(dummymodel.score(data_train, target_train)))
print("Score based on test data is\t{}".format(dummymodel.score(data_test, target_test)))

print("===============")

dummymodel2 = DummyClassifier(strategy='constant', constant=' >50K')
_ = dummymodel2.fit(data_train,target_train)


# for checking if any False is available (false referse to ' >50K')
n = 0
for i in (dummymodel2.predict(data_test) == ' >50K'):
    if i == False:
        n+=1
print("n is %d" %(n))
print("This model predicts only ' >50K'")
print("Score based on train data is\t{}".format(dummymodel2.score(data_train, target_train)))
print("Score based on test data is\t{}".format(dummymodel2.score(data_test, target_test)))
print("======== \n"
      "Let do some tweak \n"
      "Sum of scores for train data is \t %f \n"
      "Sum of scores for test data is  \t %f" %(dummymodel2.score(data_train, target_train)+dummymodel.score(data_train, target_train), dummymodel2.score(data_test, target_test)+dummymodel.score(data_test, target_test)))

😐