tomaarsen HF staff commited on
Commit
d482672
1 Parent(s): 0f08233

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -154,14 +154,14 @@ Despite originally being intended for Natural Language Inference (NLI), this dat
154
 
155
  ### `pair-score` subset
156
 
157
- * Columns: "sentence1", "sentence2", "label"
158
  * Column types: `str`, `str`, `float`
159
  * Examples:
160
  ```python
161
  {
162
  'sentence1': 'A person on a horse jumps over a broken down airplane.',
163
  'sentence2': 'A person is training his horse for a competition.',
164
- 'label': 1.0,
165
  }
166
  ```
167
  * Collection strategy: Taking the `pair-class` subset and remapping "entailment", "neutral" and "contradiction" to 1.0, 0.5 and 0.0, respectively.
 
154
 
155
  ### `pair-score` subset
156
 
157
+ * Columns: "sentence1", "sentence2", "score"
158
  * Column types: `str`, `str`, `float`
159
  * Examples:
160
  ```python
161
  {
162
  'sentence1': 'A person on a horse jumps over a broken down airplane.',
163
  'sentence2': 'A person is training his horse for a competition.',
164
+ 'score': 0.5,
165
  }
166
  ```
167
  * Collection strategy: Taking the `pair-class` subset and remapping "entailment", "neutral" and "contradiction" to 1.0, 0.5 and 0.0, respectively.