Evaluation results for caskcsg/InfoCSE-bert-base-rtd model as a base model for other tasks
#1
by
eladven
- opened
README.md
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# caskcsg/InfoCSE-bert-base-rtd model
|
| 2 |
+
This model is based on bert-base-uncased pretrained model.
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
## Model Recycling
|
| 6 |
+
|
| 7 |
+
[Evaluation on 36 datasets](https://ibm.github.io/model-recycling/model_gain_chart?avg=1.61&mnli_lp=nan&20_newsgroup=-0.93&ag_news=0.11&amazon_reviews_multi=0.62&anli=0.64&boolq=2.24&cb=7.05&cola=1.78&copa=7.55&dbpedia=-0.43&esnli=0.93&financial_phrasebank=15.77&imdb=-0.70&isear=2.44&mnli=1.03&mrpc=2.33&multirc=-3.02&poem_sentiment=9.28&qnli=1.23&qqp=0.17&rotten_tomatoes=0.98&rte=-0.05&sst2=0.69&sst_5bins=0.60&stsb=0.52&trec_coarse=1.14&trec_fine=10.87&tweet_ev_emoji=0.57&tweet_ev_emotion=1.79&tweet_ev_hate=-0.26&tweet_ev_irony=-2.58&tweet_ev_offensive=-0.95&tweet_ev_sentiment=1.42&wic=1.01&wnli=-6.90&wsc=0.38&yahoo_answers=0.68&model_name=caskcsg%2FInfoCSE-bert-base-rtd&base_name=bert-base-uncased) using caskcsg/InfoCSE-bert-base-rtd as a base model yields average score of 73.81 in comparison to 72.20 by bert-base-uncased.
|
| 8 |
+
|
| 9 |
+
The model is ranked 2nd among all tested models for the bert-base-uncased architecture as of 21/12/2022
|
| 10 |
+
Results:
|
| 11 |
+
|
| 12 |
+
| 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 |
|
| 13 |
+
|---------------:|----------:|-----------------------:|--------:|--------:|--------:|-------:|-------:|----------:|--------:|-----------------------:|-------:|--------:|--------:|--------:|----------:|-----------------:|--------:|--------:|------------------:|--------:|--------:|------------:|--------:|--------------:|------------:|-----------------:|-------------------:|----------------:|-----------------:|---------------------:|---------------------:|--------:|-------:|------:|----------------:|
|
| 14 |
+
| 82.1163 | 89.7 | 66.54 | 47.5938 | 71.1927 | 71.4286 | 83.605 | 57 | 77.7333 | 90.6352 | 84.3 | 90.88 | 71.5124 | 84.7539 | 84.3137 | 56.9513 | 75.9615 | 91.1038 | 90.4452 | 85.8349 | 59.9278 | 92.6606 | 53.3937 | 86.3785 | 97.2 | 79.2 | 36.58 | 81.703 | 52.5926 | 65.1786 | 84.4186 | 70.8971 | 64.2633 | 43.662 | 62.5 | 73 |
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
For more information, see: [Model Recycling](https://ibm.github.io/model-recycling/)
|