Inconsistent Validation results during training and by using allennlp evaluate
shmahaj opened this issue · comments
I am fine-tuning a model and the validation accuracy I get during training time using allennlp train(85%) is higher than what I get when I run allennlp evaluate(42%) on the saved model and the validation set. Could you please suggest what could be going wrong here?
This issue is being closed due to lack of activity. If you think it still needs to be addressed, please comment on this thread 👇