cardiffnlp/twitter-roberta-base-emotion

This is a roBERTa-base model trained on ~58M tweets and finetuned for emotion recognition with the TweetEval benchmark . The model is trained on 58M tweets . It is a paper that analyzes tweets using EmNLP (Findings of EMNLP 2020) The model was trained on tweets with the Twitter-roBERTA-base benchmark . It was then fed with a label mapping to map the tweets to the Tweeteval official repository . The results are based on the data from the model's training code . The code is then downloaded to the official repository of the training code and uploaded to the GitHub repository . For more information on the code, visit http://,