Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -40,7 +40,11 @@ This dataset card aims to be a base template for new datasets. It has been gener
|
|
40 |
|
41 |
### Data Fields
|
42 |
|
43 |
-
|
|
|
|
|
|
|
|
|
44 |
|
45 |
### Data Splits
|
46 |
|
|
|
40 |
|
41 |
### Data Fields
|
42 |
|
43 |
+
- `premise`: The machine translated premise in the target language.
|
44 |
+
- `hypothesis`: The machine translated premise in the target language.
|
45 |
+
- `label`: The classification label, with possible values `entailment` (0), `neutral` (1), `contradiction` (2).
|
46 |
+
- `task`: The particular NLP task that the data was drawn from (IE, IR, QA and SUM).
|
47 |
+
- `length`: The length of the text of the pair.
|
48 |
|
49 |
### Data Splits
|
50 |
|