RoBERTa Latin base model Version 2 (Uncased)
Prerequisites
transformers==4.19.2
Model architecture
This model uses RoBERTa base setttings except vocabulary size.
Tokenizer
Using BPE tokenizer with a vocabulary size 50,000.
Training Data
- Subset of CC-100/la : Monolingual Datasets from Web Crawl Data
Usage
from transformers import pipeline
unmasker = pipeline('fill-mask', model='ClassCat/roberta-base-latin-v2')
unmasker("vita brevis, ars <mask>")