Reputation: 175
Suppose we have a train dataset where the columns are bathrooms, bedrooms, sqft living, view and some other features, and price is defined there. In a Test data set all the above features are there except the price. How come we can predict the price from the data set ?
Upvotes: 0
Views: 1319
Reputation: 41
Having no label makes your test set useless so you might divide the labeled (you have called it training data here) set to train and test. This way, you will not use all your data for the training and the test portion will be used for the validation of your model.
Upvotes: 2
Reputation: 181
You can train the model with training set that you have but if you don’t have the label (ground truth) for the test set, you will not be able to validate the performance of your model (the one you have trained with training set).
Upvotes: 0