Ihor commited on
Commit
6c8daa1
1 Parent(s): dafd89b

add a remark about benchmarking

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -91,6 +91,7 @@ prob_label_is_true = probs[:,1]
91
  ```
92
 
93
  ### Benchmarking
 
94
  | Model | IMDB | AG_NEWS | Emotions |
95
  |-----------------------------|------|---------|----------|
96
  | [Bart-large-mnli (407 M)](https://huggingface.co/facebook/bart-large-mnli) | 0.89 | 0.6887 | 0.3765 |
 
91
  ```
92
 
93
  ### Benchmarking
94
+ Below, you can see the F1 score on several text classification datasets. All tested models were not fine-tuned on those datasets and were tested in a zero-shot setting.
95
  | Model | IMDB | AG_NEWS | Emotions |
96
  |-----------------------------|------|---------|----------|
97
  | [Bart-large-mnli (407 M)](https://huggingface.co/facebook/bart-large-mnli) | 0.89 | 0.6887 | 0.3765 |