Skip to content

GitLab

Switch branch/tag
Find file Normal viewHistoryPermalink
t0_eval.yaml 3.41 KB
Newer Older
lintangsutawika's avatar
can now process a benchmark that uses promptsource  
lintangsutawika committed
1 
group: t0_eval
lintangsutawika's avatar
added benchmark configs  
lintangsutawika committed
2 
task:
lintangsutawika's avatar
update and fixes  
lintangsutawika committed
3
4
5
6
7
8 
  # Coreference Resolution
  - dataset_path: super_glue
    dataset_name: wsc.fixed
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
9 
    output_type: generate_until
lintangsutawika's avatar
update and fixes  
lintangsutawika committed
10
11
12
13
14
15
16
17
18
19
20
21 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  # Coreference Resolution
  - dataset_path: winogrande
    dataset_name: winogrande_xl
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
22 
    output_type: generate_until
lintangsutawika's avatar
can now process a benchmark that uses promptsource  
lintangsutawika committed
23
24
25
26
27
28 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
lintangsutawika's avatar
update  
lintangsutawika committed
29
30
31
32
33
34 
  # Natural Language Inference
  - dataset_path: super_glue
    dataset_name: cb
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
35 
    output_type: generate_until
lintangsutawika's avatar
update  
lintangsutawika committed
36
37
38
39
40
41
42
43
44
45
46 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  - dataset_path: super_glue
    dataset_name: rte
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
47 
    output_type: generate_until
lintangsutawika's avatar
update  
lintangsutawika committed
48
49
50
51
52
53
54
55
56
57
58 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  - task: anli_r1
    dataset_path: anli
    use_prompt: promptsource:*
    training_split: train_r1
    validation_split: dev_r1
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
59 
    output_type: generate_until
lintangsutawika's avatar
update  
lintangsutawika committed
60
61
62
63
64
65
66
67
68
69
70 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  - task: anli_r2
    dataset_path: anli
    use_prompt: promptsource:*
    training_split: train_r2
    validation_split: dev_r2
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
71 
    output_type: generate_until
lintangsutawika's avatar
update  
lintangsutawika committed
72
73
74
75
76
77
78
79
80
81
82 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  - task: anli_r3
    dataset_path: anli
    use_prompt: promptsource:*
    training_split: train_r3
    validation_split: dev_r3
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
83 
    output_type: generate_until
lintangsutawika's avatar
update  
lintangsutawika committed
84
85
86
87
88
89
90
91
92
93
94
95 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  # Sentence Completion
  - dataset_path: super_glue
    dataset_name: copa
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
96 
    output_type: generate_until
lintangsutawika's avatar
update  
lintangsutawika committed
97
98
99
100
101
102 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
lintangsutawika's avatar
can now process a benchmark that uses promptsource  
lintangsutawika committed
103 
  # Natural Language Inference
lintangsutawika's avatar
update and fixes  
lintangsutawika committed
104
105
106
107 
  - dataset_path: hellaswag
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
108 
    output_type: generate_until
lintangsutawika's avatar
update and fixes  
lintangsutawika committed
109
110
111
112
113
114
115
116
117
118
119
120 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true
  # Word Sense Disambiguation
  - dataset_path: super_glue
    dataset_name: wic
    use_prompt: promptsource:*
    training_split: train
    validation_split: validation
lintangsutawika's avatar
change all mentions of `greedy_until` to `generate_until`  
lintangsutawika committed
121 
    output_type: generate_until
lintangsutawika's avatar
update and fixes  
lintangsutawika committed
122
123
124
125
126
127 
    metric_list:
      - metric: exact_match
        aggregation: mean
        higher_is_better: true
        ignore_case: true
        ignore_punctuation: true