Dylan1999/bert-finetuned-squad-accelerate model
This model is based on bert-base-cased pretrained model.
Model Recycling
Evaluation on 36 datasets using Dylan1999/bert-finetuned-squad-accelerate as a base model yields average score of 74.07 in comparison to 72.43 by bert-base-cased.
The model is ranked 2nd among all tested models for the bert-base-cased architecture as of 21/12/2022 Results:
| 20_newsgroup | ag_news | amazon_reviews_multi | anli | boolq | cb | cola | copa | dbpedia | esnli | financial_phrasebank | imdb | isear | mnli | mrpc | multirc | poem_sentiment | qnli | qqp | rotten_tomatoes | rte | sst2 | sst_5bins | stsb | trec_coarse | trec_fine | tweet_ev_emoji | tweet_ev_emotion | tweet_ev_hate | tweet_ev_irony | tweet_ev_offensive | tweet_ev_sentiment | wic | wnli | wsc | yahoo_answers |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 81.7047 | 89.1333 | 66.04 | 46.9375 | 71.0398 | 75 | 80.0575 | 55 | 79.5667 | 89.6274 | 80 | 91.044 | 69.8175 | 83.2689 | 86.2745 | 59.2822 | 73.0769 | 91.0123 | 88.9117 | 84.803 | 67.87 | 91.9725 | 50.0452 | 86.0266 | 96.4 | 82.6 | 44.214 | 79.3807 | 54.3434 | 68.2398 | 84.186 | 66.7779 | 63.4796 | 54.9296 | 63.4615 | 70.9333 |
For more information, see: Model Recycling