cardiffnlp/twitter-roberta-base-emoji

This is a roBERTa-base model trained on 58M tweets and finetuned for emoji prediction with the TweetEval benchmark . The model is trained on tweets with a dataset of 58,000 tweets . It is then fed into a model that predicts the future of emoji and emojines . The results are based on the data set of EMNLP 2020's Emojines (EMNLP-LNLP) and the dataset of tweets . The data set is based on tweets sent by Twitter's @roberta-base-base and tweets from the Twitter-roberta-base task . For example, the model was trained on over 58 million tweets to predict,