Upload tokenizer
Browse files- special_tokens_map.json +28 -0
- tokenizer.json +36 -0
special_tokens_map.json
CHANGED
@@ -20,6 +20,34 @@
|
|
20 |
"normalized": false,
|
21 |
"rstrip": false,
|
22 |
"single_word": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
}
|
24 |
],
|
25 |
"bos_token": {
|
|
|
20 |
"normalized": false,
|
21 |
"rstrip": false,
|
22 |
"single_word": false
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"content": "<stop_TMP>",
|
26 |
+
"lstrip": true,
|
27 |
+
"normalized": false,
|
28 |
+
"rstrip": false,
|
29 |
+
"single_word": false
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"content": "select",
|
33 |
+
"lstrip": true,
|
34 |
+
"normalized": false,
|
35 |
+
"rstrip": false,
|
36 |
+
"single_word": false
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"content": "from",
|
40 |
+
"lstrip": true,
|
41 |
+
"normalized": false,
|
42 |
+
"rstrip": false,
|
43 |
+
"single_word": false
|
44 |
+
},
|
45 |
+
{
|
46 |
+
"content": "left_join",
|
47 |
+
"lstrip": true,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false
|
51 |
}
|
52 |
],
|
53 |
"bos_token": {
|
tokenizer.json
CHANGED
@@ -35,6 +35,24 @@
|
|
35 |
"normalized": false,
|
36 |
"special": true
|
37 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
38 |
{
|
39 |
"id": 32000,
|
40 |
"content": "<pad>",
|
@@ -70,6 +88,24 @@
|
|
70 |
"rstrip": false,
|
71 |
"normalized": false,
|
72 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
73 |
}
|
74 |
],
|
75 |
"normalizer": {
|
|
|
35 |
"normalized": false,
|
36 |
"special": true
|
37 |
},
|
38 |
+
{
|
39 |
+
"id": 2622,
|
40 |
+
"content": "select",
|
41 |
+
"single_word": false,
|
42 |
+
"lstrip": true,
|
43 |
+
"rstrip": false,
|
44 |
+
"normalized": false,
|
45 |
+
"special": true
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"id": 3166,
|
49 |
+
"content": "from",
|
50 |
+
"single_word": false,
|
51 |
+
"lstrip": true,
|
52 |
+
"rstrip": false,
|
53 |
+
"normalized": false,
|
54 |
+
"special": true
|
55 |
+
},
|
56 |
{
|
57 |
"id": 32000,
|
58 |
"content": "<pad>",
|
|
|
88 |
"rstrip": false,
|
89 |
"normalized": false,
|
90 |
"special": true
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"id": 32004,
|
94 |
+
"content": "<stop_TMP>",
|
95 |
+
"single_word": false,
|
96 |
+
"lstrip": true,
|
97 |
+
"rstrip": false,
|
98 |
+
"normalized": false,
|
99 |
+
"special": true
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"id": 32005,
|
103 |
+
"content": "left_join",
|
104 |
+
"single_word": false,
|
105 |
+
"lstrip": true,
|
106 |
+
"rstrip": false,
|
107 |
+
"normalized": false,
|
108 |
+
"special": true
|
109 |
}
|
110 |
],
|
111 |
"normalizer": {
|