Skip to content

Code for the paper `Unsupervised Cross-Lingual Adaptation of Dependency Parsers Using CRF Autoencoders` in the findings of EMNLP 2020.

Notifications You must be signed in to change notification settings

livc/cross-crfae

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Cross CRFAE

Code for the paper Unsupervised Cross-Lingual Adaptation of Dependency Parsers Using CRF Autoencoders in the findings of EMNLP 2020. https://www.aclweb.org/anthology/2020.findings-emnlp.193/

Requirements

Usage

source:

python run.py train -p -d=$cuda -f=exp/source --feat=tag --crf --n_mlp_arc 50 --n_lstm_hidden 200 --n_lstm_layers 3 --n_embed 150 --output log/source --berts --seed 1

target (with W_Reg):

python run.py train -p -d=$cuda -f=exp/target --feat=tag --unsupervised --max_len 40 --lang ar  --n_mlp_arc 50 --n_lstm_hidden 200 --n_lstm_layers 3 --n_embed 150 --output log/target --epochs 1 --load "/path/to/cross-crfae/exp/source/model" --W_Reg --W_beta 1e8 --freeze_feat_emb --crf --bert --seed 1

About

Code for the paper `Unsupervised Cross-Lingual Adaptation of Dependency Parsers Using CRF Autoencoders` in the findings of EMNLP 2020.

Topics

Resources

Stars

Watchers

Forks

Languages