bdqnghi / infercode

[ICSE 2021] - InferCode: Self-Supervised Learning of Code Representations by Predicting Subtrees
89 stars 24 forks source link

Using F1 score or accuracy #35

Closed Avv22 closed 1 year ago

Avv22 commented 1 year ago

Hello,

If someone uses your embeddings of a source code for any downstream task your model supports, do you recommend using F1 score or accuracy to evaluate your model and why, please? In case your model F1 score is low but high accuracy, what would you recommend?