autogluon-bench/sample_configs/dataloaders/paper_image_datasets.yaml at 8b46e8a40bc27df2b9a28b12c19bde83d7e413f3 · autogluon/autogluon-bench · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
# data = custom_class(dataset_name=dataset_name, split=split, **kwargs)
# custom_class: custom dataloader class, e.g. sample_configs/vision_dataset.py


base: &base
  url: s3://zs-models/datasets/{name}.zip
  splits:
    - train
    - test
  image_columns:
    - ImageID
  text_columns:
  label_columns:
    - LabelName
  columns_to_drop:
  annotation: "{name}_{split}_annotations.csv"
  image_path: "{name}/{value}"
  metric: acc
  problem_type: multiclass

fashion_mnist:
  <<: *base
  image_path: "{split}/{value}"

casting:
  <<: *base
  image_path: "{value}"
  problem_type: binary
  metric: roc_auc

food101:
  <<: *base

oxfordflowers:
  <<: *base
  image_path: "{name}/{split}/{value}"

OxfordIIITPet:
  <<: *base
  splits:
    - train
    - validation
    - test
  annotation: "{name}_{split}_anno.csv"
  image_path: "{split}/{value}"

europeanflooddepth:
  <<: *base
  problem_type: binary
  metric: roc_auc

magnetictiledefects:
  <<: *base

stanfordcars:
  <<: *base

cub200:
  <<: *base


petfinder:
  <<: *base
  splits:
    - train
  text_columns:
    - Description
  label_columns:
    - AdoptionSpeed
  annotation: "{name}_{split}_annotations.csv"
  image_path: "{value}"
  metric: acc
  problem_type: multiclass


ham10000:
  <<: *base
  splits:
    - train
  label_columns:
    - dx
  image_path: "{split}/{value}"

cd18:
  <<: *base
  splits:
    - train
  label_columns:
    - Price
  image_path: "{split}/{value}"
  metric: rmse
  problem_type: regression


hateful_meme:
  <<: *base
  splits:
    - train
  text_columns:
    - text
  image_path: "{value}"
  metric: roc_auc
  problem_type: binary


memotion:
  <<: *base
  splits:
    - train
  text_columns:
    - text_corrected
  columns_to_drop:
    - text_ocr
  label_columns:
    - overall_sentiment
  annotation: "{name}_{split}_annotations.csv"
  image_path: "{split}/{value}"
  metric: acc
  problem_type: multiclass