anuragsingh28
commited on
Commit
·
2879607
1
Parent(s):
2a6eab3
Upload processor
Browse files- added_tokens.json +3 -10
- special_tokens_map.json +1 -8
- tokenizer.json +1 -64
added_tokens.json
CHANGED
@@ -1,12 +1,5 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
-
"<
|
5 |
-
"<s_answer>": 57526,
|
6 |
-
"<s_docvqa>": 57527,
|
7 |
-
"<s_iitcdip>": 57528,
|
8 |
-
"<s_question>": 57529,
|
9 |
-
"<s_synthdog>": 57530,
|
10 |
-
"<sep/>": 57522,
|
11 |
-
"<yes/>": 57531
|
12 |
}
|
|
|
1 |
{
|
2 |
+
"<s_iitcdip>": 57523,
|
3 |
+
"<s_synthdog>": 57524,
|
4 |
+
"<sep/>": 57522
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
}
|
special_tokens_map.json
CHANGED
@@ -1,14 +1,7 @@
|
|
1 |
{
|
2 |
"additional_special_tokens": [
|
3 |
-
"</s_answer>",
|
4 |
-
"</s_question>",
|
5 |
-
"<no/>",
|
6 |
-
"<s_answer>",
|
7 |
-
"<s_docvqa>",
|
8 |
"<s_iitcdip>",
|
9 |
-
"<
|
10 |
-
"<s_synthdog>",
|
11 |
-
"<yes/>"
|
12 |
],
|
13 |
"bos_token": "<s>",
|
14 |
"cls_token": "<s>",
|
|
|
1 |
{
|
2 |
"additional_special_tokens": [
|
|
|
|
|
|
|
|
|
|
|
3 |
"<s_iitcdip>",
|
4 |
+
"<s_synthdog>"
|
|
|
|
|
5 |
],
|
6 |
"bos_token": "<s>",
|
7 |
"cls_token": "<s>",
|
tokenizer.json
CHANGED
@@ -73,51 +73,6 @@
|
|
73 |
},
|
74 |
{
|
75 |
"id": 57523,
|
76 |
-
"content": "</s_answer>",
|
77 |
-
"single_word": false,
|
78 |
-
"lstrip": false,
|
79 |
-
"rstrip": false,
|
80 |
-
"normalized": false,
|
81 |
-
"special": true
|
82 |
-
},
|
83 |
-
{
|
84 |
-
"id": 57524,
|
85 |
-
"content": "</s_question>",
|
86 |
-
"single_word": false,
|
87 |
-
"lstrip": false,
|
88 |
-
"rstrip": false,
|
89 |
-
"normalized": false,
|
90 |
-
"special": true
|
91 |
-
},
|
92 |
-
{
|
93 |
-
"id": 57525,
|
94 |
-
"content": "<no/>",
|
95 |
-
"single_word": false,
|
96 |
-
"lstrip": false,
|
97 |
-
"rstrip": false,
|
98 |
-
"normalized": false,
|
99 |
-
"special": true
|
100 |
-
},
|
101 |
-
{
|
102 |
-
"id": 57526,
|
103 |
-
"content": "<s_answer>",
|
104 |
-
"single_word": false,
|
105 |
-
"lstrip": false,
|
106 |
-
"rstrip": false,
|
107 |
-
"normalized": false,
|
108 |
-
"special": true
|
109 |
-
},
|
110 |
-
{
|
111 |
-
"id": 57527,
|
112 |
-
"content": "<s_docvqa>",
|
113 |
-
"single_word": false,
|
114 |
-
"lstrip": false,
|
115 |
-
"rstrip": false,
|
116 |
-
"normalized": false,
|
117 |
-
"special": true
|
118 |
-
},
|
119 |
-
{
|
120 |
-
"id": 57528,
|
121 |
"content": "<s_iitcdip>",
|
122 |
"single_word": false,
|
123 |
"lstrip": false,
|
@@ -126,31 +81,13 @@
|
|
126 |
"special": true
|
127 |
},
|
128 |
{
|
129 |
-
"id":
|
130 |
-
"content": "<s_question>",
|
131 |
-
"single_word": false,
|
132 |
-
"lstrip": false,
|
133 |
-
"rstrip": false,
|
134 |
-
"normalized": false,
|
135 |
-
"special": true
|
136 |
-
},
|
137 |
-
{
|
138 |
-
"id": 57530,
|
139 |
"content": "<s_synthdog>",
|
140 |
"single_word": false,
|
141 |
"lstrip": false,
|
142 |
"rstrip": false,
|
143 |
"normalized": false,
|
144 |
"special": true
|
145 |
-
},
|
146 |
-
{
|
147 |
-
"id": 57531,
|
148 |
-
"content": "<yes/>",
|
149 |
-
"single_word": false,
|
150 |
-
"lstrip": false,
|
151 |
-
"rstrip": false,
|
152 |
-
"normalized": false,
|
153 |
-
"special": true
|
154 |
}
|
155 |
],
|
156 |
"normalizer": {
|
|
|
73 |
},
|
74 |
{
|
75 |
"id": 57523,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
"content": "<s_iitcdip>",
|
77 |
"single_word": false,
|
78 |
"lstrip": false,
|
|
|
81 |
"special": true
|
82 |
},
|
83 |
{
|
84 |
+
"id": 57524,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
85 |
"content": "<s_synthdog>",
|
86 |
"single_word": false,
|
87 |
"lstrip": false,
|
88 |
"rstrip": false,
|
89 |
"normalized": false,
|
90 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
}
|
92 |
],
|
93 |
"normalizer": {
|