Datasets:

Multilinguality:
multilingual
Size Categories:
1K<n<10K
Annotations Creators:
crowdsourced
ArXiv:
Tags:
License:
albertvillanova HF staff commited on
Commit
3c45489
1 Parent(s): dcc5275

Add X-CODAH-pl data files

Browse files
README.md CHANGED
@@ -319,13 +319,13 @@ dataset_info:
319
  dtype: string
320
  splits:
321
  - name: test
322
- num_bytes: 438824
323
  num_examples: 1000
324
  - name: validation
325
- num_bytes: 127862
326
  num_examples: 300
327
- download_size: 7519903
328
- dataset_size: 566686
329
  - config_name: X-CODAH-pt
330
  features:
331
  - name: id
@@ -975,6 +975,12 @@ configs:
975
  path: X-CODAH-nl/test-*
976
  - split: validation
977
  path: X-CODAH-nl/validation-*
 
 
 
 
 
 
978
  - config_name: X-CODAH-zh
979
  data_files:
980
  - split: test
 
319
  dtype: string
320
  splits:
321
  - name: test
322
+ num_bytes: 438538
323
  num_examples: 1000
324
  - name: validation
325
+ num_bytes: 127750
326
  num_examples: 300
327
+ download_size: 254894
328
+ dataset_size: 566288
329
  - config_name: X-CODAH-pt
330
  features:
331
  - name: id
 
975
  path: X-CODAH-nl/test-*
976
  - split: validation
977
  path: X-CODAH-nl/validation-*
978
+ - config_name: X-CODAH-pl
979
+ data_files:
980
+ - split: test
981
+ path: X-CODAH-pl/test-*
982
+ - split: validation
983
+ path: X-CODAH-pl/validation-*
984
  - config_name: X-CODAH-zh
985
  data_files:
986
  - split: test
X-CODAH-pl/test-00000-of-00001.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86037837d13ba12395b0113f544c04522d5be635e0750a3a13b3d68bf23771dd
3
+ size 194967
X-CODAH-pl/validation-00000-of-00001.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a07fca30a80a48e43b2c3ae6b249eae0fa7041478ddf87a417c707cb40bd64c3
3
+ size 59927
dataset_infos.json CHANGED
@@ -1623,53 +1623,42 @@
1623
  "features": {
1624
  "id": {
1625
  "dtype": "string",
1626
- "id": null,
1627
  "_type": "Value"
1628
  },
1629
  "lang": {
1630
  "dtype": "string",
1631
- "id": null,
1632
  "_type": "Value"
1633
  },
1634
  "question_tag": {
1635
  "dtype": "string",
1636
- "id": null,
1637
  "_type": "Value"
1638
  },
1639
  "question": {
1640
  "stem": {
1641
  "dtype": "string",
1642
- "id": null,
1643
  "_type": "Value"
1644
  },
1645
  "choices": {
1646
  "feature": {
1647
  "label": {
1648
  "dtype": "string",
1649
- "id": null,
1650
  "_type": "Value"
1651
  },
1652
  "text": {
1653
  "dtype": "string",
1654
- "id": null,
1655
  "_type": "Value"
1656
  }
1657
  },
1658
- "length": -1,
1659
- "id": null,
1660
  "_type": "Sequence"
1661
  }
1662
  },
1663
  "answerKey": {
1664
  "dtype": "string",
1665
- "id": null,
1666
  "_type": "Value"
1667
  }
1668
  },
1669
- "post_processed": null,
1670
- "supervised_keys": null,
1671
- "task_templates": null,
1672
  "builder_name": "xcsr",
 
1673
  "config_name": "X-CODAH-pl",
1674
  "version": {
1675
  "version_str": "1.1.0",
@@ -1681,27 +1670,20 @@
1681
  "splits": {
1682
  "test": {
1683
  "name": "test",
1684
- "num_bytes": 438824,
1685
  "num_examples": 1000,
1686
- "dataset_name": "xcsr"
1687
  },
1688
  "validation": {
1689
  "name": "validation",
1690
- "num_bytes": 127862,
1691
  "num_examples": 300,
1692
- "dataset_name": "xcsr"
1693
- }
1694
- },
1695
- "download_checksums": {
1696
- "https://inklab.usc.edu/XCSR/xcsr_datasets.zip": {
1697
- "num_bytes": 7519903,
1698
- "checksum": "c45b29ece740643252d5402e76be1e33f96f9d6910053f79e80d39887f10c85e"
1699
  }
1700
  },
1701
- "download_size": 7519903,
1702
- "post_processing_size": null,
1703
- "dataset_size": 566686,
1704
- "size_in_bytes": 8086589
1705
  },
1706
  "X-CODAH-pt": {
1707
  "description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n",
 
1623
  "features": {
1624
  "id": {
1625
  "dtype": "string",
 
1626
  "_type": "Value"
1627
  },
1628
  "lang": {
1629
  "dtype": "string",
 
1630
  "_type": "Value"
1631
  },
1632
  "question_tag": {
1633
  "dtype": "string",
 
1634
  "_type": "Value"
1635
  },
1636
  "question": {
1637
  "stem": {
1638
  "dtype": "string",
 
1639
  "_type": "Value"
1640
  },
1641
  "choices": {
1642
  "feature": {
1643
  "label": {
1644
  "dtype": "string",
 
1645
  "_type": "Value"
1646
  },
1647
  "text": {
1648
  "dtype": "string",
 
1649
  "_type": "Value"
1650
  }
1651
  },
 
 
1652
  "_type": "Sequence"
1653
  }
1654
  },
1655
  "answerKey": {
1656
  "dtype": "string",
 
1657
  "_type": "Value"
1658
  }
1659
  },
 
 
 
1660
  "builder_name": "xcsr",
1661
+ "dataset_name": "xcsr",
1662
  "config_name": "X-CODAH-pl",
1663
  "version": {
1664
  "version_str": "1.1.0",
 
1670
  "splits": {
1671
  "test": {
1672
  "name": "test",
1673
+ "num_bytes": 438538,
1674
  "num_examples": 1000,
1675
+ "dataset_name": null
1676
  },
1677
  "validation": {
1678
  "name": "validation",
1679
+ "num_bytes": 127750,
1680
  "num_examples": 300,
1681
+ "dataset_name": null
 
 
 
 
 
 
1682
  }
1683
  },
1684
+ "download_size": 254894,
1685
+ "dataset_size": 566288,
1686
+ "size_in_bytes": 821182
 
1687
  },
1688
  "X-CODAH-pt": {
1689
  "description": "To evaluate multi-lingual language models (ML-LMs) for commonsense reasoning in a cross-lingual zero-shot transfer setting (X-CSR), i.e., training in English and test in other languages, we create two benchmark datasets, namely X-CSQA and X-CODAH. Specifically, we automatically translate the original CSQA and CODAH datasets, which only have English versions, to 15 other languages, forming development and test sets for studying X-CSR. As our goal is to evaluate different ML-LMs in a unified evaluation protocol for X-CSR, we argue that such translated examples, although might contain noise, can serve as a starting benchmark for us to obtain meaningful analysis, before more human-translated datasets will be available in the future.\n",