bad accuracy on test data

[Caiji-123]

Hi, I find your accuracy is calculated on training data, but generally we evaluate the model on test data. I get the accuracy from test data and the result is not that good, is there anything wrong?

[Caiji-123]

the acc on test I got is 0.588 epoch 38

[plkmo]

Hi there, for semeval task the model is evaluated on test data. However for pre-training, there is no test data as the pre-training is self-supervised. You were probably looking at the code for pre-training.

Also the accuracy is evaluated on a multi-class > 2 dataset, so the baseline accuracy is not 0.5. You can use F1 for a better comparison

[Caiji-123]

已收到您的邮件