Hello! I'm exploring the XGlue evaluation module with the purpose of evaluation our custom model and found that it uses precision metric as score: https://github.com/microsoft/XGLUE/blob/master/evaluation/xglue_evaluate.py#L217 But the Readme states that POS-tagging is measured by accuracy: https://github.com/microsoft/XGLUE/blob/master/README.md#pos-tagging-pos Which metric should I use to be able to compare my results with other submission on a Leaderboard?
Hello! I'm exploring the XGlue evaluation module with the purpose of evaluation our custom model and found that it uses precision metric as score:
https://github.com/microsoft/XGLUE/blob/master/evaluation/xglue_evaluate.py#L217
But the Readme states that POS-tagging is measured by accuracy:
https://github.com/microsoft/XGLUE/blob/master/README.md#pos-tagging-pos
Which metric should I use to be able to compare my results with other submission on a Leaderboard?