-
Notifications
You must be signed in to change notification settings - Fork 16
/
Copy pathschema-token.json
58 lines (58 loc) · 2.08 KB
/
schema-token.json
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
{
"$schema": "",
"description": "A condense format for includin all token-level samples in the training dataset",
"type": "object",
"properties": {
"data": {
"type": "array",
"description": "Each element represents one sample of the model input",
"items": {
"$ref": "#/$defs/data"
}
},
"labels": {
"type": "object",
"description": "A mapping from label_id to label_name"
},
"files": {
"type": "array",
"description": "A list of strings representing the file names for the i-th object in data",
"items": {
"type": "string"
}
},
},
"$defs": {
"data": {
"type": "object",
"properties": {
"words":{
"type": "array",
"description": "A list of individual string words in this page. Empty token (\"\") should be removed."
},
"bbox": {
"type": "array",
"description": "A list of bounding boxes ([x1, y1, x2, y2]) for each token in the same order as words"
},
"labels": {
"type": "array",
"description": "A list of label_id for each token"
},
"block_ids": {
"type": "array",
"description": "[Optional] A list of ids for the corresponding block for the each token. Used for constructing block embeddings, etc",
"items": {
"type": "number"
}
},
"line_ids": {
"type": "array",
"description": "[Optional] A list of ids for the corresponding line for the each token. Used for constructing line embeddings, etc",
"items": {
"type": "number"
}
},
}
}
}
}