Skip to content

Latest commit

 

History

History
36 lines (34 loc) · 10 KB

FINETUNING_RESULT.md

File metadata and controls

36 lines (34 loc) · 10 KB
Raw

Leaderboard

Following table is a list of models with metrics fine-tuned via tweetnlp trainer. All the models can be reproduced with tweetnlp trainer (see here for tweet_eval tasks and here for others).

task language_model eval_f1 eval_f1_macro eval_accuracy link
emoji cardiffnlp/twitter-roberta-base-2021-124m 0.46 0.35 0.46 cardiffnlp/twitter-roberta-base-2021-124m-emoji
emoji cardiffnlp/twitter-roberta-base-dec2021 0.46 0.34 0.46 cardiffnlp/twitter-roberta-base-dec2021-emoji
emoji roberta-base 0.46 0.34 0.46 cardiffnlp/roberta-base-emoji
emotion cardiffnlp/twitter-roberta-base-2021-124m 0.83 0.79 0.83 cardiffnlp/twitter-roberta-base-2021-124m-emotion
emotion cardiffnlp/twitter-roberta-base-dec2021 0.85 0.82 0.85 cardiffnlp/twitter-roberta-base-dec2021-emotion
emotion roberta-base 0.82 0.79 0.82 cardiffnlp/roberta-base-emotion
hate cardiffnlp/twitter-roberta-base-2021-124m 0.56 0.53 0.56 cardiffnlp/twitter-roberta-base-2021-124m-hate
hate cardiffnlp/twitter-roberta-base-dec2021 0.57 0.54 0.57 cardiffnlp/twitter-roberta-base-dec2021-hate
hate roberta-base 0.51 0.46 0.51 cardiffnlp/roberta-base-hate
irony cardiffnlp/twitter-roberta-base-2021-124m 0.79 0.78 0.79 cardiffnlp/twitter-roberta-base-2021-124m-irony
irony cardiffnlp/twitter-roberta-base-dec2021 0.78 0.78 0.78 cardiffnlp/twitter-roberta-base-dec2021-irony
irony roberta-base 0.7 0.69 0.7 cardiffnlp/roberta-base-irony
offensive cardiffnlp/twitter-roberta-base-2021-124m 0.86 0.82 0.86 cardiffnlp/twitter-roberta-base-2021-124m-offensive
offensive cardiffnlp/twitter-roberta-base-dec2021 0.86 0.83 0.86 cardiffnlp/twitter-roberta-base-dec2021-offensive
offensive roberta-base 0.84 0.8 0.84 cardiffnlp/roberta-base-offensive
sentiment cardiffnlp/twitter-roberta-base-2021-124m 0.71 0.72 0.71 cardiffnlp/twitter-roberta-base-2021-124m-sentiment
sentiment cardiffnlp/twitter-roberta-base-dec2021 0.71 0.71 0.71 cardiffnlp/twitter-roberta-base-dec2021-sentiment
sentiment roberta-base 0.71 0.71 0.71 cardiffnlp/roberta-base-sentiment
topic_classification (single) cardiffnlp/twitter-roberta-base-2021-124m 0.9 0.8 0.9 cardiffnlp/twitter-roberta-base-2021-124m-topic-single
topic_classification (single) cardiffnlp/twitter-roberta-base-dec2021 0.9 0.79 0.9 cardiffnlp/twitter-roberta-base-dec2021-topic-single
topic_classification (single) roberta-base 0.88 0.74 0.88 cardiffnlp/roberta-base-topic-single
topic_classification (multi) cardiffnlp/twitter-roberta-base-2021-124m 0.75 0.56 0.54 cardiffnlp/twitter-roberta-base-2021-124m-topic-multi
topic_classification (multi) cardiffnlp/twitter-roberta-base-dec2021 0.75 0.56 0.53 cardiffnlp/twitter-roberta-base-dec2021-topic-multi
topic_classification (multi) roberta-base 0.75 0.6 0.53 cardiffnlp/roberta-base-topic-multi
sentiment (multilingual) bert-base-multilingual-cased 0.62 0.62 0.62 cardiffnlp/bert-base-multilingual-cased-sentiment-multilingual
sentiment (multilingual) cardiffnlp/twitter-xlm-roberta-base 0.69 0.69 0.69 cardiffnlp/twitter-xlm-roberta-base-sentiment-multilingual
sentiment (multilingual) xlm-roberta-base 0.67 0.66 0.67 cardiffnlp/xlm-roberta-base-sentiment-multilingual