gremlin97 commited on
Commit
97d9291
·
verified ·
1 Parent(s): 8b4f213

Add dataset README with metadata

Browse files
Files changed (1) hide show
  1. README.md +1 -55
README.md CHANGED
@@ -18,35 +18,6 @@ task_categories:
18
  task_ids:
19
  - multi-label-image-classification
20
  pretty_name: mars-multi-label-classification
21
- configs:
22
- - config_name: default
23
- data_files:
24
- - split: train
25
- path: data/train-*
26
- - split: val
27
- path: data/val-*
28
- - split: test
29
- path: data/test-*
30
- dataset_info:
31
- features:
32
- - name: image
33
- dtype: image
34
- - name: labels
35
- sequence: int8
36
- - name: feature_names
37
- sequence: string
38
- splits:
39
- - name: train
40
- num_bytes: 189511167.738
41
- num_examples: 1762
42
- - name: val
43
- num_bytes: 48685102.0
44
- num_examples: 443
45
- - name: test
46
- num_bytes: 78350787.0
47
- num_examples: 739
48
- download_size: 320118588
49
- dataset_size: 316547056.73800004
50
  ---
51
 
52
  # mars-multi-label-classification
@@ -96,32 +67,7 @@ The dataset contains the following classes:
96
  - **train**: 1762 images
97
  - **val**: 443 images
98
  - **test**: 739 images
99
- - **few_shot_train_10_shot**: 1310 images
100
- - **few_shot_train_15_shot**: 1357 images
101
- - **few_shot_train_1_shot**: 1198 images
102
- - **few_shot_train_20_shot**: 1402 images
103
- - **few_shot_train_2_shot**: 1212 images
104
- - **few_shot_train_5_shot**: 1249 images
105
-
106
- ## Few-shot Splits
107
-
108
- This dataset includes the following few-shot training splits:
109
-
110
- - **few_shot_train_10_shot**: 1310 images
111
- - **few_shot_train_15_shot**: 1357 images
112
- - **few_shot_train_1_shot**: 1198 images
113
- - **few_shot_train_20_shot**: 1402 images
114
- - **few_shot_train_2_shot**: 1212 images
115
- - **few_shot_train_5_shot**: 1249 images
116
-
117
- Few-shot configurations:
118
-
119
- - **10_shot.csv**
120
- - **15_shot.csv**
121
- - **1_shot.csv**
122
- - **20_shot.csv**
123
- - **2_shot.csv**
124
- - **5_shot.csv**
125
  ## Format
126
 
127
  Each example in the dataset has the following format:
 
18
  task_ids:
19
  - multi-label-image-classification
20
  pretty_name: mars-multi-label-classification
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
21
  ---
22
 
23
  # mars-multi-label-classification
 
67
  - **train**: 1762 images
68
  - **val**: 443 images
69
  - **test**: 739 images
70
+
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
71
  ## Format
72
 
73
  Each example in the dataset has the following format: