Datasets:
Remove config names as yaml keys (#4367)
Browse files* validate YAML part
* fix bad YAML keys
* don't have config names as yaml keys
* style
* Remove hacking license tags when mirroring datasets on the Hub
* fix test
* fix style
* more fixes
* remove dots in QA column mapping keys
Commit from https://github.com/huggingface/datasets/commit/3f3013441e4af9bf4efa752b5e6e184a588b7a80
README.md
CHANGED
|
@@ -11,18 +11,14 @@ language_creators:
|
|
| 11 |
languages:
|
| 12 |
- en
|
| 13 |
licenses:
|
| 14 |
-
- cc-by-4
|
| 15 |
multilinguality:
|
| 16 |
- monolingual
|
| 17 |
size_categories:
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
squad:
|
| 23 |
-
- n<1K
|
| 24 |
-
trex:
|
| 25 |
-
- 1M<n<10M
|
| 26 |
source_datasets:
|
| 27 |
- extended|conceptnet5
|
| 28 |
- extended|squad
|
|
@@ -34,6 +30,11 @@ task_ids:
|
|
| 34 |
- text-classification-other-probing
|
| 35 |
- text-scoring
|
| 36 |
paperswithcode_id: lama
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
---
|
| 38 |
|
| 39 |
# Dataset Card for LAMA: LAnguage Model Analysis - a dataset for probing and analyzing the factual and commonsense knowledge contained in pretrained language models.
|
|
|
|
| 11 |
languages:
|
| 12 |
- en
|
| 13 |
licenses:
|
| 14 |
+
- cc-by-4.0
|
| 15 |
multilinguality:
|
| 16 |
- monolingual
|
| 17 |
size_categories:
|
| 18 |
+
- 10K<n<100K
|
| 19 |
+
- 1K<n<10K
|
| 20 |
+
- 1M<n<10M
|
| 21 |
+
- n<1K
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
source_datasets:
|
| 23 |
- extended|conceptnet5
|
| 24 |
- extended|squad
|
|
|
|
| 30 |
- text-classification-other-probing
|
| 31 |
- text-scoring
|
| 32 |
paperswithcode_id: lama
|
| 33 |
+
configs:
|
| 34 |
+
- conceptnet
|
| 35 |
+
- google_re
|
| 36 |
+
- squad
|
| 37 |
+
- trex
|
| 38 |
---
|
| 39 |
|
| 40 |
# Dataset Card for LAMA: LAnguage Model Analysis - a dataset for probing and analyzing the factual and commonsense knowledge contained in pretrained language models.
|