Evaluation results for jfarmerphd/bert-finetuned-squad-accelerate model as a base model for other tasks
Browse filesAs part of a research effort to identify high quality models in Huggingface that can serve as base models for further finetuning, we evaluated this by finetuning on 36 datasets. The model ranks 3rd among all tested models for the bert-base-cased architecture as of 09/01/2023.
To share this information with others in your model card, please add the following evaluation results to your README.md page.
For more information please see https://ibm.github.io/model-recycling/ or contact me.
Best regards,
Elad Venezian
eladv@il.ibm.com
IBM Research AI
README.md
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# jfarmerphd/bert-finetuned-squad-accelerate model
|
| 2 |
+
This model is based on bert-base-cased pretrained model.
|
| 3 |
+
|
| 4 |
+
|
| 5 |
+
## Model Recycling
|
| 6 |
+
|
| 7 |
+
[Evaluation on 36 datasets](https://ibm.github.io/model-recycling/model_gain_chart?avg=1.62&mnli_lp=nan&20_newsgroup=-0.65&ag_news=-0.33&amazon_reviews_multi=0.13&anli=0.84&boolq=3.17&cb=7.95&cola=-0.26&copa=-2.15&dbpedia=-1.14&esnli=-0.13&financial_phrasebank=14.24&imdb=-0.13&isear=1.04&mnli=-0.33&mrpc=5.31&multirc=0.12&poem_sentiment=5.38&qnli=0.98&qqp=-0.09&rotten_tomatoes=0.35&rte=6.68&sst2=0.37&sst_5bins=0.04&stsb=1.04&trec_coarse=0.37&trec_fine=7.42&tweet_ev_emoji=-0.22&tweet_ev_emotion=-1.08&tweet_ev_hate=0.62&tweet_ev_irony=2.66&tweet_ev_offensive=0.98&tweet_ev_sentiment=0.27&wic=0.58&wnli=4.01&wsc=1.54&yahoo_answers=-1.19&model_name=jfarmerphd%2Fbert-finetuned-squad-accelerate&base_name=bert-base-cased) using jfarmerphd/bert-finetuned-squad-accelerate as a base model yields average score of 74.05 in comparison to 72.43 by bert-base-cased.
|
| 8 |
+
|
| 9 |
+
The model is ranked 3rd among all tested models for the bert-base-cased architecture as of 09/01/2023
|
| 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 |
+
| 81.094 | 88.7333 | 65.84 | 47.4062 | 71.4373 | 71.4286 | 81.5916 | 50 | 77.6333 | 89.5053 | 82.6 | 91.012 | 69.4263 | 83.0553 | 88.2353 | 60.5817 | 73.0769 | 90.9757 | 89.8541 | 84.8968 | 69.3141 | 91.8578 | 51.448 | 85.562 | 97 | 80.4 | 44.018 | 77.7621 | 53.4007 | 67.8571 | 85.2326 | 68.4956 | 65.3605 | 56.338 | 63.4615 | 69.8333 |
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
For more information, see: [Model Recycling](https://ibm.github.io/model-recycling/)
|