Hugging Face's logo

Datasets:
THUDM
/
VisionRewardDB-Video

Modalities:
Text
Formats:
parquet
Size:
10K - 100K
ArXiv:
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files Community
2
CodeZzz commited on
Commit
6ca43e1
·
1 Parent(s): 211d8dd

update

Browse files
Files changed (1) hide show
  1. README.md +124 -108
README.md CHANGED
@@ -1,121 +1,137 @@
1
  ---
2
- license: apache-2.0
3
-
4
  dataset_info:
 
5
  features:
6
- train:
7
- - name: video
8
- dtype: binary
9
- - name: internal_id
10
- dtype: string
11
- - name: prompt
12
- dtype: string
13
- - name: url
14
- dtype: string
15
- - name: annotation
16
- struct:
17
- - name: alignment
18
- dtype: int64
19
- range: [1,5]
20
- - name: composition
21
- dtype: int64
22
- range: [1,3]
23
- - name: focus
24
- dtype: int64
25
- range: [1,3]
26
- - name: camera movement
27
- dtype: int64
28
- range: [1,3]
29
- - name: color
30
- dtype: int64
31
- range: [1,5]
32
- - name: lighting accurate
33
- dtype: int64
34
- range: [1,4]
35
- - name: lighting aes
36
- dtype: int64
37
- range: [1,5]
38
- - name: shape at beginning
39
- dtype: int64
40
- range: [0,3]
41
- - name: shape throughout
42
- dtype: int64
43
- range: [0,4]
44
- - name: object motion dynamic
45
- dtype: int64
46
- range: [1,5]
47
- - name: camera motion dynamic
48
- dtype: int64
49
- range: [1,5]
50
- - name: movement smoothness
51
- dtype: int64
52
- range: [0,4]
53
- - name: movement reality
54
- dtype: int64
55
- range: [0,4]
56
- - name: clear
57
- dtype: int64
58
- range: [1,5]
59
- - name: image quality stability
60
- dtype: int64
61
- range: [1,5]
62
- - name: camera stability
63
- dtype: int64
64
- range: [1,3]
65
- - name: detail refinement
66
- dtype: int64
67
- range: [1,5]
68
- - name: letters
69
- dtype: int64
70
- range: [1,4]
71
- - name: physics law
72
- dtype: int64
73
- range: [1,5]
74
- - name: unsafe type # Not used for training
75
- dtype: int64
76
- range: [1,5]
77
- - name: safety
78
- dtype: int64
79
- range: [1,5]
80
- - name: meta_result
81
- sequence:
82
- dtype: int64
83
- - name: meta_mask
84
- sequence:
85
- dtype: int64
86
-
87
-
88
- regression:
89
- - name: internal_id
90
- dtype: string
91
- - name: prompt
92
- dtype: string
93
- - name: standard_answer
94
- dtype: string
95
- - name: file1
96
- dtype: binary
97
- - name: file2
98
- dtype: binary
99
-
100
- monetbench:
101
- - name: internal_id
102
- dtype: string
103
- - name: prompt
104
- dtype: string
105
- - name: standard_answer
106
- dtype: string
107
- - name: file1
108
- dtype: binary
109
- - name: file2
110
- dtype: binary
111
- config_name: default
112
  splits:
113
  - name: train
114
  num_examples: 40743
 
 
 
 
 
 
 
 
 
 
 
 
 
 
115
  - name: regression
116
  num_examples: 2260
 
 
 
 
 
 
 
 
 
 
 
 
 
 
117
  - name: monetbench
118
  num_examples: 1000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
119
  ---
120
 
121
  # VisionRewardDB-Video
 
1
  ---
 
 
2
  dataset_info:
3
+ - config_name: train
4
  features:
5
+ - name: video
6
+ dtype: binary
7
+ - name: internal_id
8
+ dtype: string
9
+ - name: prompt
10
+ dtype: string
11
+ - name: url
12
+ dtype: string
13
+ - name: annotation
14
+ struct:
15
+ - name: alignment
16
+ dtype: int64
17
+ range: [1,5]
18
+ - name: composition
19
+ dtype: int64
20
+ range: [1,3]
21
+ - name: focus
22
+ dtype: int64
23
+ range: [1,3]
24
+ - name: camera movement
25
+ dtype: int64
26
+ range: [1,3]
27
+ - name: color
28
+ dtype: int64
29
+ range: [1,5]
30
+ - name: lighting accurate
31
+ dtype: int64
32
+ range: [1,4]
33
+ - name: lighting aes
34
+ dtype: int64
35
+ range: [1,5]
36
+ - name: shape at beginning
37
+ dtype: int64
38
+ range: [0,3]
39
+ - name: shape throughout
40
+ dtype: int64
41
+ range: [0,4]
42
+ - name: object motion dynamic
43
+ dtype: int64
44
+ range: [1,5]
45
+ - name: camera motion dynamic
46
+ dtype: int64
47
+ range: [1,5]
48
+ - name: movement smoothness
49
+ dtype: int64
50
+ range: [0,4]
51
+ - name: movement reality
52
+ dtype: int64
53
+ range: [0,4]
54
+ - name: clear
55
+ dtype: int64
56
+ range: [1,5]
57
+ - name: image quality stability
58
+ dtype: int64
59
+ range: [1,5]
60
+ - name: camera stability
61
+ dtype: int64
62
+ range: [1,3]
63
+ - name: detail refinement
64
+ dtype: int64
65
+ range: [1,5]
66
+ - name: letters
67
+ dtype: int64
68
+ range: [1,4]
69
+ - name: physics law
70
+ dtype: int64
71
+ range: [1,5]
72
+ - name: unsafe type
73
+ dtype: int64
74
+ range: [1,5]
75
+ - name: safety
76
+ dtype: int64
77
+ range: [1,5]
78
+ - name: meta_result
79
+ sequence:
80
+ dtype: int64
81
+ - name: meta_mask
82
+ sequence:
83
+ dtype: int64
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
84
  splits:
85
  - name: train
86
  num_examples: 40743
87
+
88
+ - config_name: regression
89
+ features:
90
+ - name: internal_id
91
+ dtype: string
92
+ - name: prompt
93
+ dtype: string
94
+ - name: standard_answer
95
+ dtype: string
96
+ - name: file1
97
+ dtype: binary
98
+ - name: file2
99
+ dtype: binary
100
+ splits:
101
  - name: regression
102
  num_examples: 2260
103
+
104
+ - config_name: monetbench
105
+ features:
106
+ - name: internal_id
107
+ dtype: string
108
+ - name: prompt
109
+ dtype: string
110
+ - name: standard_answer
111
+ dtype: string
112
+ - name: file1
113
+ dtype: binary
114
+ - name: file2
115
+ dtype: binary
116
+ splits:
117
  - name: monetbench
118
  num_examples: 1000
119
+
120
+ configs:
121
+ - config_name: train
122
+ data_files:
123
+ - split: train
124
+ path: data/train-*
125
+ - config_name: regression
126
+ data_files:
127
+ - split: regression
128
+ path: regression/train-*
129
+ - config_name: monetbench
130
+ data_files:
131
+ - split: monetbench
132
+ path: monetbench/test-*
133
+
134
+ license: apache-2.0
135
  ---
136
 
137
  # VisionRewardDB-Video