The evaluation metric for this competition is [Macro F1-Score](https://en.wikipedia.org/wiki/F-score). The F1 score, commonly used in information retrieval, 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 `tp` to all actual positives `tp+fn`. The F1 score is given by:
$$F1 = 2{p\.r \over p\+r}$$
$$where, p = {tp \over tp\+fp}, r = {tp \over tp\+fn}$$

The F1 metric weights recall and precision equally, and a good retrieval algorithm will maximize both precision and recall simultaneously. Thus, moderately good performance on both will be favored over extremely good performance on one and poor performance on the other.
Macro F1-score calculates the F1 score for each class in the dataset and takes the average which is unweighted.

## Submission File
For each 'id' in the test set, you must assign a sentiment score on the scale of 1-5. The file should contain a header and have the following format:

  `id`,`sentiment`
    684,5
    715,1
    552,3
    etc.

**Good Luck!** We wish all participants the best of luck in the challenge. May the sharpest detectives help Anya in her quest to revive her cosmic companion, Gojo, by correctly interpreting the reviews.