Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -90,16 +90,16 @@ carbonio-14 O
|
|
| 90 |
Factuality:
|
| 91 |
|
| 92 |
- Fine-grained
|
| 93 |
-
- CERTAINTY: {CERTAIN, PROBABLE, POSSIBLE, UNDERSPECIFIED}
|
| 94 |
-
- POLARITY: {POSITIVE, NEGATIVE, UNDERSPECIFIED}
|
| 95 |
-
- TIME: {PRESENT/PAST, FUTURE, UNDERSPECIFIED}
|
| 96 |
|
| 97 |
- Coarse-grained
|
| 98 |
-
- {FACTUAL, NON-FACTUAL, COUNTERFACTUAL, UNDERSPECIFIED}
|
| 99 |
|
| 100 |
|
| 101 |
Modality:
|
| 102 |
-
- {WILL, FINAL, CONCESSIVE, POSSIBILITY, CAPABILITY, DUTY, COERCION, EXHORTATIVE, COMMITMENT, DECISION}
|
| 103 |
|
| 104 |
|
| 105 |
### Data format
|
|
@@ -115,7 +115,6 @@ For the sake of reproducibility, we provide, for each configuration, the 5 folds
|
|
| 115 |
The data split follows a 60/20/20 ratio and has been created in a stratified way. This means each train/dev/test set contains (approx) the same relative distribution of classes.
|
| 116 |
|
| 117 |
|
| 118 |
-
|
| 119 |
## Additional Information
|
| 120 |
|
| 121 |
|
|
|
|
| 90 |
Factuality:
|
| 91 |
|
| 92 |
- Fine-grained
|
| 93 |
+
- CERTAINTY: {`CERTAIN`, `PROBABLE`, `POSSIBLE`, `UNDERSPECIFIED`}
|
| 94 |
+
- POLARITY: {`POSITIVE`, `NEGATIVE`, `UNDERSPECIFIED`}
|
| 95 |
+
- TIME: {`PRESENT/PAST`, `FUTURE`, `UNDERSPECIFIED`}
|
| 96 |
|
| 97 |
- Coarse-grained
|
| 98 |
+
- {`FACTUAL`, `NON-FACTUAL`, `COUNTERFACTUAL`, `UNDERSPECIFIED`}
|
| 99 |
|
| 100 |
|
| 101 |
Modality:
|
| 102 |
+
- {`WILL`, `FINAL`, `CONCESSIVE`, `POSSIBILITY`, `CAPABILITY`, `DUTY`, `COERCION`, `EXHORTATIVE`, `COMMITMENT`, `DECISION`}
|
| 103 |
|
| 104 |
|
| 105 |
### Data format
|
|
|
|
| 115 |
The data split follows a 60/20/20 ratio and has been created in a stratified way. This means each train/dev/test set contains (approx) the same relative distribution of classes.
|
| 116 |
|
| 117 |
|
|
|
|
| 118 |
## Additional Information
|
| 119 |
|
| 120 |
|