Resolved: Practice Exam - Question 1
Dear Professor Hristina
I am having difficulty getting the class prediction for question 1, I am writing the code as follows however I am not getting the right answer, could you explain to me how I should call the predict function to get the right classes for the model?
x_train, y_train = make_blobs(n_samples = [50]*4,
n_features = 2,
centers = [(-1, 1),(-2,5), (4, 0),(3,6)],
cluster_std = 2,
random_state = 365)
x_test, y_test = make_blobs(n_samples = [40]*4,
n_features = 2,
centers = [(-1, 1),(-2,5), (4, 0),(3,6)],
cluster_std = 2,
random_state = 365)
data = [x_train,y_train]
data = pd.DataFrame(data = x_train, columns = ['year', 'price'])
data['Target'] = y_train
data
clf = KNeighborsClassifier(n_neighbors = 5, weights = 'uniform')
clf.fit(x_train, y_train)
predictions = clf.predict(x_test)
prediction[0]
Hey Mani,
Thank you for reaching out and for engaging with the Machine Learning with KNN course!
Your code looks great! Make sure that you change the n_neighbors and the weights parameters in the following line of code
clf = KNeighborsClassifier(n_neighbors = 5, weights = 'uniform')
based on the model. First, change the parameters to 3 neighbors and uniform weights, then, to 5 neighbors and uniform weights, and lastly, to 5 neighbors and distance weights. For each of these cases, you should get a prediction matching the correct answer.
Hope this helps! Let me know if you still encounter issues with this question.
Kind regards,
365 Hristina
