Evaluation results for ibm/ColD-Fusion 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 1st among all tested models for the roberta-base architecture as of 21/12/2022.
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
CHANGED
@@ -51,6 +51,20 @@ output = model(encoded_input)
|
|
51 |
```
|
52 |
|
53 |
## Evaluation results
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
54 |
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
|
55 |
When fine-tuned on downstream tasks, this model achieves the following results:
|
56 |
|
|
|
51 |
```
|
52 |
|
53 |
## Evaluation results
|
54 |
+
|
55 |
+
## Model Recycling
|
56 |
+
|
57 |
+
[Evaluation on 36 datasets](https://ibm.github.io/model-recycling/model_gain_chart?avg=2.25&mnli_lp=nan&20_newsgroup=0.54&ag_news=0.03&amazon_reviews_multi=-0.32&anli=1.59&boolq=2.68&cb=19.73&cola=-0.22&copa=23.30&dbpedia=1.34&esnli=0.15&financial_phrasebank=2.99&imdb=-0.04&isear=1.06&mnli=0.31&mrpc=-0.86&multirc=2.50&poem_sentiment=1.63&qnli=-0.00&qqp=0.40&rotten_tomatoes=3.41&rte=12.80&sst2=1.30&sst_5bins=-0.30&stsb=1.38&trec_coarse=-0.11&trec_fine=2.64&tweet_ev_emoji=0.00&tweet_ev_emotion=1.22&tweet_ev_hate=1.55&tweet_ev_irony=6.37&tweet_ev_offensive=1.38&tweet_ev_sentiment=-0.60&wic=3.17&wnli=-6.90&wsc=-2.69&yahoo_answers=-0.53&model_name=ibm%2FColD-Fusion&base_name=roberta-base) using ibm/ColD-Fusion as a base model yields average score of 78.47 in comparison to 76.22 by roberta-base.
|
58 |
+
|
59 |
+
The model is ranked 1st among all tested models for the roberta-base architecture as of 21/12/2022
|
60 |
+
Results:
|
61 |
+
|
62 |
+
| 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 |
|
63 |
+
|---------------:|----------:|-----------------------:|--------:|--------:|-----:|--------:|-------:|----------:|--------:|-----------------------:|-------:|--------:|--------:|--------:|----------:|-----------------:|--------:|-------:|------------------:|--------:|--------:|------------:|--------:|--------------:|------------:|-----------------:|-------------------:|----------------:|-----------------:|---------------------:|---------------------:|-------:|--------:|--------:|----------------:|
|
64 |
+
| 85.8205 | 89.8 | 66.26 | 51.9375 | 81.3761 | 87.5 | 83.3174 | 72 | 78.6333 | 91.1441 | 88.1 | 93.864 | 73.5332 | 87.2966 | 87.0098 | 63.717 | 85.5769 | 92.4034 | 91.113 | 91.8386 | 85.1986 | 95.4128 | 56.3801 | 91.2964 | 97 | 90.4 | 46.306 | 83.0401 | 54.4444 | 77.9337 | 85.9302 | 70.4331 | 68.652 | 47.8873 | 60.5769 | 71.8667 |
|
65 |
+
|
66 |
+
|
67 |
+
For more information, see: [Model Recycling](https://ibm.github.io/model-recycling/)
|
68 |
See full evaluation results of this model and many more [here](https://ibm.github.io/model-recycling/roberta-base_table.html)
|
69 |
When fine-tuned on downstream tasks, this model achieves the following results:
|
70 |
|