Skip to content

LivNLP/multi-source-attention

Repository files navigation

Structure

load_data.py --- helper method to get train and test data under Blitzer2007

utils.py --- helper method to get train and test data under Chen2012

self_train.py --- learn pseudo classifier from source domain data

domain_similarity.py --- learn relatedness map between instances from sources and target

exam_stars.py --- evaluation on different sorting methods for pseudo labelled instacnes selection

sorting_methods.py --- choose best selection methods

att_model.py --- train domain attention model

multi_processor.py --- helper method n multi-processing for attention model

evidence.py --- generate evidences for label prediction

To Run

Follow the arguments in the program

About

Multi-Source Domain Attention

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages