JeremiahZ autoevaluator HF staff commited on
Commit
e7669ea
1 Parent(s): c2f58c0

Add evaluation results on the qnli config of glue (#1)

Browse files

- Add evaluation results on the qnli config of glue (5183b2d12646ef546a8a367c65ebfe5e785f2107)


Co-authored-by: Evaluation Bot <autoevaluator@users.noreply.huggingface.co>

Files changed (1) hide show
  1. README.md +33 -0
README.md CHANGED
@@ -22,6 +22,39 @@ model-index:
22
  - name: Accuracy
23
  type: accuracy
24
  value: 0.9245835621453414
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
22
  - name: Accuracy
23
  type: accuracy
24
  value: 0.9245835621453414
25
+ - task:
26
+ type: natural-language-inference
27
+ name: Natural Language Inference
28
+ dataset:
29
+ name: glue
30
+ type: glue
31
+ config: qnli
32
+ split: validation
33
+ metrics:
34
+ - name: Accuracy
35
+ type: accuracy
36
+ value: 0.924400512538898
37
+ verified: true
38
+ - name: Precision
39
+ type: precision
40
+ value: 0.9171997157071784
41
+ verified: true
42
+ - name: Recall
43
+ type: recall
44
+ value: 0.9348062296269467
45
+ verified: true
46
+ - name: AUC
47
+ type: auc
48
+ value: 0.9744865501321541
49
+ verified: true
50
+ - name: F1
51
+ type: f1
52
+ value: 0.9259192825112107
53
+ verified: true
54
+ - name: loss
55
+ type: loss
56
+ value: 0.2990749478340149
57
+ verified: true
58
  ---
59
 
60
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You