MoritzLaurer HF staff commited on
Commit
c72e58a
1 Parent(s): 074d9b5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -2
README.md CHANGED
@@ -52,12 +52,17 @@ training_args = TrainingArguments(
52
  ### Eval results
53
  The model was evaluated using the test sets for MultiNLI and ANLI and the dev set for Fever-NLI
54
  dataset | accuracy
55
- -------|---------
56
- mnli_m_mm | 0.903/0.903
 
57
  fever-nli | 0.777
58
  anli-all | 0.579
59
  anli-r3 | 0.495
60
 
 
 
 
 
61
 
62
  ## Limitations and bias
63
  Please consult the original DeBERTa paper and literature on different NLI datasets for potential biases.
 
52
  ### Eval results
53
  The model was evaluated using the test sets for MultiNLI and ANLI and the dev set for Fever-NLI
54
  dataset | accuracy
55
+ -------|------
56
+ mnli-m | 0.903
57
+ mnli-mm | 0.903
58
  fever-nli | 0.777
59
  anli-all | 0.579
60
  anli-r3 | 0.495
61
 
62
+ metric | | precision | recall | accuracy (not balanced)
63
+ -------|---------|----------|---------|----------
64
+ accuracy | 0.773 | 0.772 | 0.771 | 0.771
65
+
66
 
67
  ## Limitations and bias
68
  Please consult the original DeBERTa paper and literature on different NLI datasets for potential biases.