search

macanv / BERT-BiLSTM-CRF-NER

Tensorflow solution of NER task Using BiLSTM-CRF model with Google BERT Fine-tuning And private Server services
https://github.com/macanv/BERT-BiLSMT-CRF-NER
4.67k stars 1.25k forks source link

What's the purpose of using a concated linear transformation in lstm_crf_layer() #363

Open real-brilliant opened 3 years ago

real-brilliant commented 3 years ago

https://github.com/macanv/BERT-BiLSTM-CRF-NER/blob/master/bert_base/train/lstm_crf_layer.py#L105

Since they functioned as one linear transformation, what's the purpose to split it into two operations? decrease trainable params?