Skip to content

Code for running the experiments in Deep Subjecthood: Higher Order Grammatical Features in Multilingual BERT

Notifications You must be signed in to change notification settings

toizzy/deep-subjecthood

Repository files navigation

Deep Subjecthood: Higher Order Grammatical Features in Multilingual BERT

This is the code used to run the experiments in Deep Subjecthood: Higher Order Grammatical Features in Multilingual BERT.

Use

run_one_experiment.py runs one experiment: training a classifier to classify subjects from objects in mBERT embeddings in one training language, and testing this classifier on one test language. Classifiers and mBERT embeddings are cached, to reuse for other experiments with the same train/test language.

make_script_to_run_all_experiments.py makes a script to run run_one_experiment.py for every pair of languages in source_langs.txt and sink_langs.txt

To reproduce the experiments in the paper:

  1. Change the paths in source_langs.txt and sink_langs.txt, so that they match the location of the Universal Dependency Treebanks on your machine
  2. Run make_script_to_run_all_experiments.py to create the scripts that will run the experiments. This will create a script run_batch_0.sh (or more if you're running for multiple seeds
  3. Run run_batch_0.sh

About

Code for running the experiments in Deep Subjecthood: Higher Order Grammatical Features in Multilingual BERT

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages