add tokenizer
Browse files- tokenizer.json +21 -9
- vocab.txt +5 -5
tokenizer.json
CHANGED
@@ -1,7 +1,19 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
@@ -186,17 +198,17 @@
|
|
186 |
"r": 56,
|
187 |
"s": 57,
|
188 |
"##\"": 58,
|
189 |
-
"##
|
190 |
-
"##
|
191 |
-
"##
|
192 |
"##S": 62,
|
193 |
"##E": 63,
|
194 |
"##P": 64,
|
195 |
"##]": 65,
|
196 |
-
"##
|
197 |
-
"##
|
198 |
-
"##
|
199 |
-
"##
|
200 |
}
|
201 |
}
|
202 |
}
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": {
|
4 |
+
"direction": "Right",
|
5 |
+
"max_length": 1000,
|
6 |
+
"strategy": "LongestFirst",
|
7 |
+
"stride": 0
|
8 |
+
},
|
9 |
+
"padding": {
|
10 |
+
"strategy": "BatchLongest",
|
11 |
+
"direction": "Right",
|
12 |
+
"pad_to_multiple_of": null,
|
13 |
+
"pad_id": 0,
|
14 |
+
"pad_type_id": 0,
|
15 |
+
"pad_token": "[PAD]"
|
16 |
+
},
|
17 |
"added_tokens": [
|
18 |
{
|
19 |
"id": 0,
|
|
|
198 |
"r": 56,
|
199 |
"s": 57,
|
200 |
"##\"": 58,
|
201 |
+
"##[": 59,
|
202 |
+
"##C": 60,
|
203 |
+
"##O": 61,
|
204 |
"##S": 62,
|
205 |
"##E": 63,
|
206 |
"##P": 64,
|
207 |
"##]": 65,
|
208 |
+
"##B": 66,
|
209 |
+
"##F": 67,
|
210 |
+
"##N": 68,
|
211 |
+
"##c": 69
|
212 |
}
|
213 |
}
|
214 |
}
|
vocab.txt
CHANGED
@@ -57,14 +57,14 @@ o
|
|
57 |
r
|
58 |
s
|
59 |
##"
|
60 |
-
##N
|
61 |
-
##F
|
62 |
##[
|
|
|
|
|
63 |
##S
|
64 |
##E
|
65 |
##P
|
66 |
##]
|
67 |
-
##O
|
68 |
-
##c
|
69 |
-
##C
|
70 |
##B
|
|
|
|
|
|
|
|
57 |
r
|
58 |
s
|
59 |
##"
|
|
|
|
|
60 |
##[
|
61 |
+
##C
|
62 |
+
##O
|
63 |
##S
|
64 |
##E
|
65 |
##P
|
66 |
##]
|
|
|
|
|
|
|
67 |
##B
|
68 |
+
##F
|
69 |
+
##N
|
70 |
+
##c
|