Upload 4 files
Browse files- added_tokens.json +2 -5
- special_tokens_map.json +2 -17
- tokenizer_config.json +4 -31
added_tokens.json
CHANGED
|
@@ -99,9 +99,6 @@
|
|
| 99 |
"<extra_id_98>": 32001,
|
| 100 |
"<extra_id_99>": 32000,
|
| 101 |
"<extra_id_9>": 32090,
|
| 102 |
-
"
|
| 103 |
-
"
|
| 104 |
-
"category:": 32103,
|
| 105 |
-
"classifiction:": 32102,
|
| 106 |
-
"generation:": 32104
|
| 107 |
}
|
|
|
|
| 99 |
"<extra_id_98>": 32001,
|
| 100 |
"<extra_id_99>": 32000,
|
| 101 |
"<extra_id_9>": 32090,
|
| 102 |
+
"article: ": 32100,
|
| 103 |
+
"category: ": 32101
|
|
|
|
|
|
|
|
|
|
| 104 |
}
|
special_tokens_map.json
CHANGED
|
@@ -1,35 +1,20 @@
|
|
| 1 |
{
|
| 2 |
"additional_special_tokens": [
|
| 3 |
{
|
| 4 |
-
"content": "article:",
|
| 5 |
"lstrip": false,
|
| 6 |
"normalized": false,
|
| 7 |
"rstrip": false,
|
| 8 |
"single_word": false
|
| 9 |
},
|
| 10 |
{
|
| 11 |
-
"content": "
|
| 12 |
-
"lstrip": false,
|
| 13 |
-
"normalized": false,
|
| 14 |
-
"rstrip": false,
|
| 15 |
-
"single_word": false
|
| 16 |
-
},
|
| 17 |
-
{
|
| 18 |
-
"content": "category:",
|
| 19 |
-
"lstrip": false,
|
| 20 |
-
"normalized": false,
|
| 21 |
-
"rstrip": false,
|
| 22 |
-
"single_word": false
|
| 23 |
-
},
|
| 24 |
-
{
|
| 25 |
-
"content": "generation:",
|
| 26 |
"lstrip": false,
|
| 27 |
"normalized": false,
|
| 28 |
"rstrip": false,
|
| 29 |
"single_word": false
|
| 30 |
}
|
| 31 |
],
|
| 32 |
-
"bos_token": "<s>",
|
| 33 |
"eos_token": {
|
| 34 |
"content": "</s>",
|
| 35 |
"lstrip": false,
|
|
|
|
| 1 |
{
|
| 2 |
"additional_special_tokens": [
|
| 3 |
{
|
| 4 |
+
"content": "article: ",
|
| 5 |
"lstrip": false,
|
| 6 |
"normalized": false,
|
| 7 |
"rstrip": false,
|
| 8 |
"single_word": false
|
| 9 |
},
|
| 10 |
{
|
| 11 |
+
"content": "category: ",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
"lstrip": false,
|
| 13 |
"normalized": false,
|
| 14 |
"rstrip": false,
|
| 15 |
"single_word": false
|
| 16 |
}
|
| 17 |
],
|
|
|
|
| 18 |
"eos_token": {
|
| 19 |
"content": "</s>",
|
| 20 |
"lstrip": false,
|
tokenizer_config.json
CHANGED
|
@@ -825,7 +825,7 @@
|
|
| 825 |
"special": true
|
| 826 |
},
|
| 827 |
"32100": {
|
| 828 |
-
"content": "
|
| 829 |
"lstrip": false,
|
| 830 |
"normalized": false,
|
| 831 |
"rstrip": false,
|
|
@@ -833,31 +833,7 @@
|
|
| 833 |
"special": true
|
| 834 |
},
|
| 835 |
"32101": {
|
| 836 |
-
"content": "
|
| 837 |
-
"lstrip": false,
|
| 838 |
-
"normalized": false,
|
| 839 |
-
"rstrip": false,
|
| 840 |
-
"single_word": false,
|
| 841 |
-
"special": true
|
| 842 |
-
},
|
| 843 |
-
"32102": {
|
| 844 |
-
"content": "classifiction:",
|
| 845 |
-
"lstrip": false,
|
| 846 |
-
"normalized": false,
|
| 847 |
-
"rstrip": false,
|
| 848 |
-
"single_word": false,
|
| 849 |
-
"special": true
|
| 850 |
-
},
|
| 851 |
-
"32103": {
|
| 852 |
-
"content": "category:",
|
| 853 |
-
"lstrip": false,
|
| 854 |
-
"normalized": false,
|
| 855 |
-
"rstrip": false,
|
| 856 |
-
"single_word": false,
|
| 857 |
-
"special": true
|
| 858 |
-
},
|
| 859 |
-
"32104": {
|
| 860 |
-
"content": "generation:",
|
| 861 |
"lstrip": false,
|
| 862 |
"normalized": false,
|
| 863 |
"rstrip": false,
|
|
@@ -866,12 +842,9 @@
|
|
| 866 |
}
|
| 867 |
},
|
| 868 |
"additional_special_tokens": [
|
| 869 |
-
"article:",
|
| 870 |
-
"
|
| 871 |
-
"category:",
|
| 872 |
-
"generation:"
|
| 873 |
],
|
| 874 |
-
"bos_token": "<s>",
|
| 875 |
"clean_up_tokenization_spaces": true,
|
| 876 |
"eos_token": "</s>",
|
| 877 |
"extra_ids": 100,
|
|
|
|
| 825 |
"special": true
|
| 826 |
},
|
| 827 |
"32100": {
|
| 828 |
+
"content": "article: ",
|
| 829 |
"lstrip": false,
|
| 830 |
"normalized": false,
|
| 831 |
"rstrip": false,
|
|
|
|
| 833 |
"special": true
|
| 834 |
},
|
| 835 |
"32101": {
|
| 836 |
+
"content": "category: ",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 837 |
"lstrip": false,
|
| 838 |
"normalized": false,
|
| 839 |
"rstrip": false,
|
|
|
|
| 842 |
}
|
| 843 |
},
|
| 844 |
"additional_special_tokens": [
|
| 845 |
+
"article: ",
|
| 846 |
+
"category: "
|
|
|
|
|
|
|
| 847 |
],
|
|
|
|
| 848 |
"clean_up_tokenization_spaces": true,
|
| 849 |
"eos_token": "</s>",
|
| 850 |
"extra_ids": 100,
|