lhoestq HF staff commited on
Commit
a32ffe4
1 Parent(s): e86ed2f

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

Files changed (1) hide show
  1. README.md +22 -81
README.md CHANGED
@@ -6,30 +6,13 @@ language_creators:
6
  languages:
7
  - en
8
  licenses:
9
- - cc-by-sa-4-0
10
  multilinguality:
11
  - monolingual
12
  size_categories:
13
- dyda_da:
14
- - 100K<n<1M
15
- dyda_e:
16
- - 100K<n<1M
17
- iemocap:
18
- - 10K<n<100K
19
- maptask:
20
- - 10K<n<100K
21
- meld_e:
22
- - 10K<n<100K
23
- meld_s:
24
- - 10K<n<100K
25
- mrda:
26
- - 100K<n<1M
27
- oasis:
28
- - 10K<n<100K
29
- sem:
30
- - 1K<n<10K
31
- swda:
32
- - 100K<n<1M
33
  source_datasets:
34
  - original
35
  task_categories:
@@ -37,68 +20,26 @@ task_categories:
37
  - fill-mask
38
  - text-classification
39
  task_ids:
40
- dyda_da:
41
- - text-scoring
42
- - dialogue-modeling
43
- - language-modeling
44
- - masked-language-modeling
45
- - text-classification-other-dialogue-act-classification
46
- dyda_e:
47
- - text-scoring
48
- - dialogue-modeling
49
- - language-modeling
50
- - masked-language-modeling
51
- - text-classification-other-emotion-classification
52
- iemocap:
53
- - text-scoring
54
- - dialogue-modeling
55
- - language-modeling
56
- - masked-language-modeling
57
- - text-classification-other-emotion-classification
58
- maptask:
59
- - text-scoring
60
- - dialogue-modeling
61
- - language-modeling
62
- - masked-language-modeling
63
- - text-classification-other-dialogue-act-classification
64
- meld_e:
65
- - text-scoring
66
- - dialogue-modeling
67
- - language-modeling
68
- - masked-language-modeling
69
- - text-classification-other-emotion-classification
70
- meld_s:
71
- - text-scoring
72
- - dialogue-modeling
73
- - language-modeling
74
- - masked-language-modeling
75
- - sentiment-classification
76
- mrda:
77
- - text-scoring
78
- - dialogue-modeling
79
- - language-modeling
80
- - masked-language-modeling
81
- - text-classification-other-dialogue-act-classification
82
- oasis:
83
- - text-scoring
84
- - dialogue-modeling
85
- - language-modeling
86
- - masked-language-modeling
87
- - text-classification-other-dialogue-act-classification
88
- sem:
89
- - text-scoring
90
- - dialogue-modeling
91
- - language-modeling
92
- - masked-language-modeling
93
- - sentiment-classification
94
- swda:
95
- - text-scoring
96
- - dialogue-modeling
97
- - language-modeling
98
- - masked-language-modeling
99
- - text-classification-other-dialogue-act-classification
100
  paperswithcode_id: null
101
  pretty_name: SILICONE Benchmark
 
 
 
 
 
 
 
 
 
 
 
102
  ---
103
 
104
  # Dataset Card for SILICONE Benchmark
 
6
  languages:
7
  - en
8
  licenses:
9
+ - cc-by-sa-4.0
10
  multilinguality:
11
  - monolingual
12
  size_categories:
13
+ - 100K<n<1M
14
+ - 10K<n<100K
15
+ - 1K<n<10K
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  source_datasets:
17
  - original
18
  task_categories:
 
20
  - fill-mask
21
  - text-classification
22
  task_ids:
23
+ - dialogue-modeling
24
+ - language-modeling
25
+ - masked-language-modeling
26
+ - sentiment-classification
27
+ - text-classification-other-dialogue-act-classification
28
+ - text-classification-other-emotion-classification
29
+ - text-scoring
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30
  paperswithcode_id: null
31
  pretty_name: SILICONE Benchmark
32
+ configs:
33
+ - dyda_da
34
+ - dyda_e
35
+ - iemocap
36
+ - maptask
37
+ - meld_e
38
+ - meld_s
39
+ - mrda
40
+ - oasis
41
+ - sem
42
+ - swda
43
  ---
44
 
45
  # Dataset Card for SILICONE Benchmark