Datasets:
Add details Spanish test set
Browse files
README.md
CHANGED
@@ -422,6 +422,8 @@ The SemRel2024 dataset covers the following 14 languages:
|
|
422 |
13. Spanish (_esp_)
|
423 |
14. Telugu (_tel_)
|
424 |
|
|
|
|
|
425 |
## Dataset Structure
|
426 |
|
427 |
### Data Instances
|
@@ -430,18 +432,15 @@ Each instance in the dataset consists of two text segments and a relatedness sco
|
|
430 |
|
431 |
```
|
432 |
{
|
433 |
-
"
|
434 |
-
"
|
435 |
-
"
|
436 |
}
|
437 |
-
|
438 |
-
tweet: a string feature.
|
439 |
-
label: a classification label, with possible values including positive, negative and neutral.
|
440 |
```
|
441 |
|
442 |
-
-
|
443 |
-
-
|
444 |
-
-
|
445 |
|
446 |
|
447 |
## Citation Information
|
|
|
422 |
13. Spanish (_esp_)
|
423 |
14. Telugu (_tel_)
|
424 |
|
425 |
+
**Note**: Spanish test labels are all -1 because the Spanish team retained the gold test labels to avoid contamination problems in future benchmarking. We refer to the [CodaLab contest website](https://codalab.lisn.upsaclay.fr/competitions/15715) to evaluate your predictions, which will remain open.
|
426 |
+
|
427 |
## Dataset Structure
|
428 |
|
429 |
### Data Instances
|
|
|
432 |
|
433 |
```
|
434 |
{
|
435 |
+
"sentence1": "string",
|
436 |
+
"sentence2": "string",
|
437 |
+
"label": float
|
438 |
}
|
|
|
|
|
|
|
439 |
```
|
440 |
|
441 |
+
- sentence1: a string feature representing the first text segment.
|
442 |
+
- sentence2: a string feature representing the second text segment.
|
443 |
+
- label: a float value representing the semantic relatedness score between sentence1 and sentence2, typically ranging from 0 (not related at all) to 1 (highly related).
|
444 |
|
445 |
|
446 |
## Citation Information
|