How to tweak the learning rate of CRF layer?

allanj / pytorch_neural_crf

Pytorch implementation of LSTM/BERT-CRF for named entity recognition

359 stars 62 forks source link

How to tweak the learning rate of CRF layer? #26

Closed Hansyvea closed 3 years ago

allanj commented 3 years ago

In short, you can have a two optimization group (when creating the optimizer), one is other parameters, another is the parameters in the CRF layer (i.e., transition)