Use the sample_submission.csv file to store all your predicted values and then upload that file as your prediction file.
Download the data file. You are given a dataset with 24 attributes and 1031 tuples. We have also provided a final Class column for 175 tuples to give you some reference to match your cluster labels with the expected class labels.
You need to build a model using clustering only that can classify the given tuples into two classes.
The evaluation metric for this competition is Accuracy Score.
Submission Format
For every id in the unlabelled dataset i.e. 856 tuples
, submission files should contain two columns: id and Class. id must be exaclty similar to the one provided in the train dataset.