Ihor commited on
Commit
381e69d
1 Parent(s): a4554c8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -0
README.md CHANGED
@@ -14,6 +14,7 @@ tags:
14
  pretty_name: biotechnews
15
  size_categories:
16
  - n<1K
 
17
  ---
18
 
19
  ### Key aspects
@@ -67,6 +68,19 @@ The dataset consists of **31** classes, including None values.
67
  * funding round - raising a new round of funding.
68
  * patent publication - publication of a new patent filing.
69
 
 
 
 
 
 
 
 
 
 
 
 
 
 
70
 
71
  ### Feedback
72
  We value your input! Share your feedback and suggestions to help us improve our models and datasets.
 
14
  pretty_name: biotechnews
15
  size_categories:
16
  - n<1K
17
+ license: odc-by
18
  ---
19
 
20
  ### Key aspects
 
68
  * funding round - raising a new round of funding.
69
  * patent publication - publication of a new patent filing.
70
 
71
+ ### Benchmark
72
+ We trained various models with binary-cross entropy loss and evaluated them on the test set.
73
+
74
+ | Model | Accuracy | F1 | Precision | Recall |
75
+ |-----------------|----------|-------|-----------|--------|
76
+ | DeBERTa-small | 96.58 | 67.69 | 74.18 | 62.19 |
77
+ | DeBERTa-base | 96.60 | 67.55 | 74.81 | 61.58 |
78
+ | DeBERTa-large | 96.99 | 74.07 | 73.46 | 74.69 |
79
+ | SciBERT-uncased | 96.57 | 68.07 | 73.07 | 63.71 |
80
+ | Flan-T5-base | 96.85 | 71.10 | 75.71 | 67.07 |
81
+
82
+ ### Recommended reading:
83
+ [Finally, a decent multi-label classification benchmark is created: a prominent zero-shot dataset.](https://medium.com/p/4d90c9e1c718)
84
 
85
  ### Feedback
86
  We value your input! Share your feedback and suggestions to help us improve our models and datasets.