zhuqi

The loss will increase after reducing to about 10

1

I train the model on Ontonotes5.0 chinese dataset, at first it works fine with constantly decreasing loss, but after 30 epoches, the loss will increase after reducing to about 10....

Have anyone tried using the code to train the model on ontonote5.0-chinese dataset? dose it work? the loss will down and then constantly rise after some steps, looking forward to...