File size: 2,696 Bytes
e89eee3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
# Dylan1999/bert-finetuned-squad-accelerate model
This model is based on bert-base-cased pretrained model.
## Model Recycling
[Evaluation on 36 datasets](https://ibm.github.io/model-recycling/model_gain_chart?avg=1.64&mnli_lp=nan&20_newsgroup=-0.03&ag_news=0.07&amazon_reviews_multi=0.33&anli=0.37&boolq=2.77&cb=11.52&cola=-1.79&copa=2.85&dbpedia=0.80&esnli=-0.01&financial_phrasebank=11.64&imdb=-0.10&isear=1.43&mnli=-0.12&mrpc=3.35&multirc=-1.18&poem_sentiment=5.38&qnli=1.02&qqp=-1.04&rotten_tomatoes=0.26&rte=5.23&sst2=0.48&sst_5bins=-1.36&stsb=1.51&trec_coarse=-0.23&trec_fine=9.62&tweet_ev_emoji=-0.03&tweet_ev_emotion=0.54&tweet_ev_hate=1.57&tweet_ev_irony=3.04&tweet_ev_offensive=-0.06&tweet_ev_sentiment=-1.45&wic=-1.30&wnli=2.61&wsc=1.54&yahoo_answers=-0.09&model_name=Dylan1999%2Fbert-finetuned-squad-accelerate&base_name=bert-base-cased) 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](https://ibm.github.io/model-recycling/)
|