aravind-selvam
commited on
Commit
•
c5d5234
1
Parent(s):
a87af08
Upload processor (#3)
Browse files- Upload processor (6174ddc6b8a85c177cadd34815bbcd64a6f16fb5)
- added_tokens.json +12 -15
- preprocessor_config.json +2 -2
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +4 -8
- tokenizer.json +14 -41
added_tokens.json
CHANGED
@@ -1,20 +1,17 @@
|
|
1 |
{
|
2 |
-
"</s_Chart_title>":
|
3 |
-
"</s_X_label>":
|
4 |
-
"</
|
5 |
-
"</
|
6 |
-
"</
|
7 |
-
"
|
8 |
-
"
|
9 |
-
"<
|
10 |
-
"<
|
11 |
-
"<s_Y_label>":
|
|
|
12 |
"<s_iitcdip>": 57523,
|
13 |
"<s_synthdog>": 57524,
|
14 |
-
"<s_x_ticks>": 57531,
|
15 |
-
"<s_x_title>": 57529,
|
16 |
-
"<s_y_ticks>": 57527,
|
17 |
-
"<s_y_title>": 57525,
|
18 |
"<sep/>": 57522,
|
19 |
-
"<xt>":
|
20 |
}
|
|
|
1 |
{
|
2 |
+
"</s_Chart_title>": 57534,
|
3 |
+
"</s_X_label>": 57532,
|
4 |
+
"</s_X_ticks>": 57530,
|
5 |
+
"</s_Y_label>": 57528,
|
6 |
+
"</s_Y_ticks>": 57526,
|
7 |
+
"<one>": 57536,
|
8 |
+
"<s_Chart_title>": 57533,
|
9 |
+
"<s_X_label>": 57531,
|
10 |
+
"<s_X_ticks>": 57529,
|
11 |
+
"<s_Y_label>": 57527,
|
12 |
+
"<s_Y_ticks>": 57525,
|
13 |
"<s_iitcdip>": 57523,
|
14 |
"<s_synthdog>": 57524,
|
|
|
|
|
|
|
|
|
15 |
"<sep/>": 57522,
|
16 |
+
"<xt>": 57535
|
17 |
}
|
preprocessor_config.json
CHANGED
@@ -20,7 +20,7 @@
|
|
20 |
"resample": 2,
|
21 |
"rescale_factor": 0.00392156862745098,
|
22 |
"size": {
|
23 |
-
"height":
|
24 |
-
"width":
|
25 |
}
|
26 |
}
|
|
|
20 |
"resample": 2,
|
21 |
"rescale_factor": 0.00392156862745098,
|
22 |
"size": {
|
23 |
+
"height": 800,
|
24 |
+
"width": 800
|
25 |
}
|
26 |
}
|
sentencepiece.bpe.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cb9e3dce4c326195d08fc3dd0f7e2eee1da8595c847bf4c1a9c78b7a82d47e2d
|
3 |
+
size 1296245
|
special_tokens_map.json
CHANGED
@@ -1,15 +1,11 @@
|
|
1 |
{
|
2 |
"additional_special_tokens": [
|
3 |
-
"<
|
4 |
-
"</
|
5 |
-
"<s_y_ticks>",
|
6 |
-
"</s_y_ticks>",
|
7 |
-
"<s_x_title>",
|
8 |
-
"</s_x_title>",
|
9 |
-
"<s_x_ticks>",
|
10 |
-
"</s_x_ticks>",
|
11 |
"<s_Y_label>",
|
12 |
"</s_Y_label>",
|
|
|
|
|
13 |
"<s_X_label>",
|
14 |
"</s_X_label>",
|
15 |
"<s_Chart_title>",
|
|
|
1 |
{
|
2 |
"additional_special_tokens": [
|
3 |
+
"<s_Y_ticks>",
|
4 |
+
"</s_Y_ticks>",
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"<s_Y_label>",
|
6 |
"</s_Y_label>",
|
7 |
+
"<s_X_ticks>",
|
8 |
+
"</s_X_ticks>",
|
9 |
"<s_X_label>",
|
10 |
"</s_X_label>",
|
11 |
"<s_Chart_title>",
|
tokenizer.json
CHANGED
@@ -77,7 +77,7 @@
|
|
77 |
},
|
78 |
{
|
79 |
"id": 57525,
|
80 |
-
"content": "<
|
81 |
"single_word": false,
|
82 |
"lstrip": false,
|
83 |
"rstrip": false,
|
@@ -86,7 +86,7 @@
|
|
86 |
},
|
87 |
{
|
88 |
"id": 57526,
|
89 |
-
"content": "</
|
90 |
"single_word": false,
|
91 |
"lstrip": false,
|
92 |
"rstrip": false,
|
@@ -95,7 +95,7 @@
|
|
95 |
},
|
96 |
{
|
97 |
"id": 57527,
|
98 |
-
"content": "<
|
99 |
"single_word": false,
|
100 |
"lstrip": false,
|
101 |
"rstrip": false,
|
@@ -104,7 +104,7 @@
|
|
104 |
},
|
105 |
{
|
106 |
"id": 57528,
|
107 |
-
"content": "</
|
108 |
"single_word": false,
|
109 |
"lstrip": false,
|
110 |
"rstrip": false,
|
@@ -113,7 +113,7 @@
|
|
113 |
},
|
114 |
{
|
115 |
"id": 57529,
|
116 |
-
"content": "<
|
117 |
"single_word": false,
|
118 |
"lstrip": false,
|
119 |
"rstrip": false,
|
@@ -122,7 +122,7 @@
|
|
122 |
},
|
123 |
{
|
124 |
"id": 57530,
|
125 |
-
"content": "</
|
126 |
"single_word": false,
|
127 |
"lstrip": false,
|
128 |
"rstrip": false,
|
@@ -131,7 +131,7 @@
|
|
131 |
},
|
132 |
{
|
133 |
"id": 57531,
|
134 |
-
"content": "<
|
135 |
"single_word": false,
|
136 |
"lstrip": false,
|
137 |
"rstrip": false,
|
@@ -140,7 +140,7 @@
|
|
140 |
},
|
141 |
{
|
142 |
"id": 57532,
|
143 |
-
"content": "</
|
144 |
"single_word": false,
|
145 |
"lstrip": false,
|
146 |
"rstrip": false,
|
@@ -149,7 +149,7 @@
|
|
149 |
},
|
150 |
{
|
151 |
"id": 57533,
|
152 |
-
"content": "<
|
153 |
"single_word": false,
|
154 |
"lstrip": false,
|
155 |
"rstrip": false,
|
@@ -158,7 +158,7 @@
|
|
158 |
},
|
159 |
{
|
160 |
"id": 57534,
|
161 |
-
"content": "</
|
162 |
"single_word": false,
|
163 |
"lstrip": false,
|
164 |
"rstrip": false,
|
@@ -167,7 +167,7 @@
|
|
167 |
},
|
168 |
{
|
169 |
"id": 57535,
|
170 |
-
"content": "<
|
171 |
"single_word": false,
|
172 |
"lstrip": false,
|
173 |
"rstrip": false,
|
@@ -176,39 +176,12 @@
|
|
176 |
},
|
177 |
{
|
178 |
"id": 57536,
|
179 |
-
"content": "
|
180 |
-
"single_word": false,
|
181 |
-
"lstrip": false,
|
182 |
-
"rstrip": false,
|
183 |
-
"normalized": false,
|
184 |
-
"special": true
|
185 |
-
},
|
186 |
-
{
|
187 |
-
"id": 57537,
|
188 |
-
"content": "<s_Chart_title>",
|
189 |
"single_word": false,
|
190 |
"lstrip": false,
|
191 |
"rstrip": false,
|
192 |
-
"normalized":
|
193 |
-
"special":
|
194 |
-
},
|
195 |
-
{
|
196 |
-
"id": 57538,
|
197 |
-
"content": "</s_Chart_title>",
|
198 |
-
"single_word": false,
|
199 |
-
"lstrip": false,
|
200 |
-
"rstrip": false,
|
201 |
-
"normalized": false,
|
202 |
-
"special": true
|
203 |
-
},
|
204 |
-
{
|
205 |
-
"id": 57539,
|
206 |
-
"content": "<xt>",
|
207 |
-
"single_word": false,
|
208 |
-
"lstrip": false,
|
209 |
-
"rstrip": false,
|
210 |
-
"normalized": false,
|
211 |
-
"special": true
|
212 |
}
|
213 |
],
|
214 |
"normalizer": {
|
|
|
77 |
},
|
78 |
{
|
79 |
"id": 57525,
|
80 |
+
"content": "<s_Y_ticks>",
|
81 |
"single_word": false,
|
82 |
"lstrip": false,
|
83 |
"rstrip": false,
|
|
|
86 |
},
|
87 |
{
|
88 |
"id": 57526,
|
89 |
+
"content": "</s_Y_ticks>",
|
90 |
"single_word": false,
|
91 |
"lstrip": false,
|
92 |
"rstrip": false,
|
|
|
95 |
},
|
96 |
{
|
97 |
"id": 57527,
|
98 |
+
"content": "<s_Y_label>",
|
99 |
"single_word": false,
|
100 |
"lstrip": false,
|
101 |
"rstrip": false,
|
|
|
104 |
},
|
105 |
{
|
106 |
"id": 57528,
|
107 |
+
"content": "</s_Y_label>",
|
108 |
"single_word": false,
|
109 |
"lstrip": false,
|
110 |
"rstrip": false,
|
|
|
113 |
},
|
114 |
{
|
115 |
"id": 57529,
|
116 |
+
"content": "<s_X_ticks>",
|
117 |
"single_word": false,
|
118 |
"lstrip": false,
|
119 |
"rstrip": false,
|
|
|
122 |
},
|
123 |
{
|
124 |
"id": 57530,
|
125 |
+
"content": "</s_X_ticks>",
|
126 |
"single_word": false,
|
127 |
"lstrip": false,
|
128 |
"rstrip": false,
|
|
|
131 |
},
|
132 |
{
|
133 |
"id": 57531,
|
134 |
+
"content": "<s_X_label>",
|
135 |
"single_word": false,
|
136 |
"lstrip": false,
|
137 |
"rstrip": false,
|
|
|
140 |
},
|
141 |
{
|
142 |
"id": 57532,
|
143 |
+
"content": "</s_X_label>",
|
144 |
"single_word": false,
|
145 |
"lstrip": false,
|
146 |
"rstrip": false,
|
|
|
149 |
},
|
150 |
{
|
151 |
"id": 57533,
|
152 |
+
"content": "<s_Chart_title>",
|
153 |
"single_word": false,
|
154 |
"lstrip": false,
|
155 |
"rstrip": false,
|
|
|
158 |
},
|
159 |
{
|
160 |
"id": 57534,
|
161 |
+
"content": "</s_Chart_title>",
|
162 |
"single_word": false,
|
163 |
"lstrip": false,
|
164 |
"rstrip": false,
|
|
|
167 |
},
|
168 |
{
|
169 |
"id": 57535,
|
170 |
+
"content": "<xt>",
|
171 |
"single_word": false,
|
172 |
"lstrip": false,
|
173 |
"rstrip": false,
|
|
|
176 |
},
|
177 |
{
|
178 |
"id": 57536,
|
179 |
+
"content": "<one>",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
180 |
"single_word": false,
|
181 |
"lstrip": false,
|
182 |
"rstrip": false,
|
183 |
+
"normalized": true,
|
184 |
+
"special": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
185 |
}
|
186 |
],
|
187 |
"normalizer": {
|