regrad evaluation

[Astudnew]

Hello, thank you for sharing this work. Regarding the evaluation of LCF- ATEPC model why do you just use F1 and accuracy for evaluation why yo do not use Class-wise performance why do not use class-wise precision, recall, and F1 score? can we use these performance metrics with LCF-ATEPC Thanks

[yangheng95]

You can try setting config.show_metric=True

[Astudnew]

Thanks but these metrics (class-wise precision, recall, and F1 score) are important for LCF-ATEPC performance evaluation (for ATE and APC)? or F1 and Accuracy are sufficient as in your paper( A multi-task learning model for Chinese-oriented aspect polarity classification and aspect term extraction)?

[yangheng95]

Actually it depends on what you think, important or not