The evaluation metric for this competition is F1-Score. The F1- Score, commonly used in bianry classification, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (tp) to all predicted positives (tp + fp). Recall is the ratio of true positives to all actual positives (tp + fn). The F1 score is given by:
\[ F1 = 2\frac{p \cdot r}{p+r}\ \ \mathrm{where}\ \ p = \frac{tp}{tp+fp},\ \ r = \frac{tp}{tp+fn} \]
Submission Format
For every sample in the dataset
, submission files should contain two columns: id and label in each row.  These two columns should be separated by a comma.
The file should contain a header and have the following format:
id,label
458,1
You can find a sample submission file (sample_Submission.csv) in the Data page.
Leader Board
The leader board includes two parts: one is public leader board which is visible to you and the other is private leader board. For this competition, we set all testing data as private leader board. So you can not see your score on public leader board. You must split your training data to hold out a validation dataset to prevent overfitting. You can choose two submissions for the evaluation for the private leader board. You can hand-select the eligible submissions, or will otherwise default to a random scoring submissions for the private leader board evaluation. **Your final results are determined by the results on the private leader board.**