Skip to content

globalise-huygens/GloBERTise

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GloBERTise: a bespoke encoder LLM for the Dutch East India Company (VOC) Archives

Usage

This is a Transformer-based encoder Language Model pre-trained on the masked language modelling objective. It is a RoBERTa architecture. The model was pre-trained from scratch on Early Modern Dutch from the Dutch East India Company Archives, digitized with a Handwritten Text Recognition Module. There was no other data involved in pre-training. The model can be used to further develop NLP technology to be used in software that supports research into the archive. In other words, the model can be fine-tuned for any NLP task, but it is built to be fine-tuned on tasks in the domain of Early Modern VOC Dutch. Download our model here.

Data

The data used for pre-training is publicly available on Dataverse. It comprises a little over 6GB in text data after pre-processing.

Development

Four models were pre-trained. There were two sets of parameter settings, varying in num_training_steps and num_warmup_steps which influence the optimizer settings (AdamW). I pre-trained two seperate models per set of paramater settings in order to investgate stochasticity between initializations. For all four models, these were some of the most important parameter settings (static for all models)

param setting
learning rate 0.0003
betas [ 0.9, 0.98]
weight_decay 0.01
num_train_epochs 2
per_device_train_batch_size 40
gradient_accumulation_steps 10
fp16 true

All parameter settings can be found per model on our Huggingface (in files).

Acknowledgements

The models were pre-trained on the Snellius HPC cluster, funded by Surf and developed within the GLOBALISE project, funded by NWO under project number 175.2019.003

Code

This code was repurposed from pre-training MedRoBERTa.nl

Evaluation

To evaluate the model on binary event detection, see my other repo

More info

About

Repo describing the creation of a domain-specific Transformer Language Model (encoder) for the archives of the Dutch East India Company

Resources

Stars

2 stars

Watchers

2 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages