Philipp Schmid • 5/22/2020

BERT Text Classification in a different language

This technical tutorial explains how to build a monolingual, non-English multi-class text classification model using BERT. It covers the selection of a pre-trained model, using the Simple Transformers library, and fine-tuning on the Germeval 2019 dataset of German tweets to classify abusive language. The guide includes steps for installation, training, evaluation, and making predictions.

0 comments

#Transformers #NLP #Text Classification