language: en tags: - emotion-classification datasets: - bdotloh/empathetic-dialogues-contexts
This is a distilbert-base-uncased model fine-tuned for the purpose of classifying (emotional) contexts in the Empathetic Dialogues dataset.
Limitations and bias
- Unable to ascertain the degree of cultural specificity for the context that a respondent described when given an emotion label (i.e., p(description | emotion, culture))
Top 1 accuracy: 53.4 Top 5 accuracy: 86.1