search

mandarjoshi90 / coref

BERT for Coreference Resolution
Apache License 2.0
441 stars 92 forks source link

Model structure redundancy #64

Open grig-guz opened 4 years ago

grig-guz commented 4 years ago

Hi,

The span width embedding over here: https://github.com/mandarjoshi90/coref/blob/bd04f2e19b9dcc0b8bba848a335e4af3be50741c/independent.py#L379 is pretty much equivalent to the span embedding over there, since the width embedding is concatenated to other span embeddings and then passed through a linear layer: https://github.com/mandarjoshi90/coref/blob/bd04f2e19b9dcc0b8bba848a335e4af3be50741c/independent.py#L362 I am trying to reimplement your model in Pytorch, so I was just wondering if there is any rationale for using two sets of span width embeddings?

Thank you.

Fantabulous-J commented 4 years ago

Hi @grig-guz! I have also implemented this model using Pytorch but always have a performance gap of around 1.2 F1 scores with the official results reported on paper. How does your implementation go on? Maybe we could share some ideas and experiences with each other.

grig-guz commented 4 years ago

Hi @Fantabulous-J, sure. I've got around 74 F1 on the dev set with Spanbert-Base, didn't run on the test set yet. My email is on my github page, you can write me there.