Upload 6 files
Browse files- config.json +207 -0
- merges.txt +0 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +59 -0
- vocab.json +0 -0
config.json
ADDED
@@ -0,0 +1,207 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "kunalr63/roberta-retrained",
|
3 |
+
"architectures": [
|
4 |
+
"RobertaForTokenClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"id2label": {
|
14 |
+
"0": "O",
|
15 |
+
"1": "B-NAME",
|
16 |
+
"2": "L-NAME",
|
17 |
+
"3": "B-C_DESIG",
|
18 |
+
"4": "I-C_DESIG",
|
19 |
+
"5": "L-C_DESIG",
|
20 |
+
"6": "B-CANDIDATE_ADDRESS",
|
21 |
+
"7": "I-CANDIDATE_ADDRESS",
|
22 |
+
"8": "L-CANDIDATE_ADDRESS",
|
23 |
+
"9": "B-PHONE",
|
24 |
+
"10": "I-PHONE",
|
25 |
+
"11": "L-PHONE",
|
26 |
+
"12": "U-EMAIL",
|
27 |
+
"13": "U-DOB",
|
28 |
+
"14": "B-T_EXP",
|
29 |
+
"15": "I-T_EXP",
|
30 |
+
"16": "L-T_EXP",
|
31 |
+
"17": "B-EXP_COMPANY_WISE",
|
32 |
+
"18": "I-EXP_COMPANY_WISE",
|
33 |
+
"19": "L-EXP_COMPANY_WISE",
|
34 |
+
"20": "B-PROJECTS",
|
35 |
+
"21": "I-PROJECTS",
|
36 |
+
"22": "L-PROJECTS",
|
37 |
+
"23": "B-SKILLS",
|
38 |
+
"24": "L-SKILLS",
|
39 |
+
"25": "U-SKILLS",
|
40 |
+
"26": "I-SKILLS",
|
41 |
+
"27": "B-CERTIFICATE",
|
42 |
+
"28": "I-CERTIFICATE",
|
43 |
+
"29": "L-CERTIFICATE",
|
44 |
+
"30": "B-G_IN",
|
45 |
+
"31": "I-G_IN",
|
46 |
+
"32": "L-G_IN",
|
47 |
+
"33": "B-G_CLG",
|
48 |
+
"34": "I-G_CLG",
|
49 |
+
"35": "L-G_CLG",
|
50 |
+
"36": "U-G_YEAR",
|
51 |
+
"37": "U-PHONE",
|
52 |
+
"38": "B-HIGH_FROM",
|
53 |
+
"39": "L-HIGH_FROM",
|
54 |
+
"40": "B-HIGH_PER",
|
55 |
+
"41": "L-HIGH_PER",
|
56 |
+
"42": "B-INTER_FROM",
|
57 |
+
"43": "L-INTER_FROM",
|
58 |
+
"44": "B-INTER_PER",
|
59 |
+
"45": "L-INTER_PER",
|
60 |
+
"46": "B-GRAD_PER",
|
61 |
+
"47": "L-GRAD_PER",
|
62 |
+
"48": "B-DOB",
|
63 |
+
"49": "I-DOB",
|
64 |
+
"50": "L-DOB",
|
65 |
+
"51": "U-NAME",
|
66 |
+
"52": "B-EMAIL",
|
67 |
+
"53": "I-EMAIL",
|
68 |
+
"54": "L-EMAIL",
|
69 |
+
"55": "I-INTER_FROM",
|
70 |
+
"56": "I-HIGH_FROM",
|
71 |
+
"57": "I-NAME",
|
72 |
+
"58": "B-PG_FROM",
|
73 |
+
"59": "L-PG_FROM",
|
74 |
+
"60": "B-PG_IN",
|
75 |
+
"61": "I-PG_IN",
|
76 |
+
"62": "L-PG_IN",
|
77 |
+
"63": "U-PG_YEAR",
|
78 |
+
"64": "U-G_IN",
|
79 |
+
"65": "I-PG_FROM",
|
80 |
+
"66": "U-CANDIDATE_ADDRESS",
|
81 |
+
"67": "U-G_CLG",
|
82 |
+
"68": "U-PG_IN",
|
83 |
+
"69": "B-PG_YEAR",
|
84 |
+
"70": "I-PG_YEAR",
|
85 |
+
"71": "L-PG_YEAR",
|
86 |
+
"72": "B-G_YEAR",
|
87 |
+
"73": "I-G_YEAR",
|
88 |
+
"74": "L-G_YEAR",
|
89 |
+
"75": "U-PG_FROM",
|
90 |
+
"76": "U-INTER_FROM",
|
91 |
+
"77": "U-HIGH_FROM",
|
92 |
+
"78": "U-GRAD_PER",
|
93 |
+
"79": "U-INTER_PER",
|
94 |
+
"80": "U-HIGH_PER",
|
95 |
+
"81": "U-T_EXP",
|
96 |
+
"82": "I-GRAD_PER",
|
97 |
+
"83": "U-PROJECTS",
|
98 |
+
"84": "U-CERTIFICATE",
|
99 |
+
"85": "U-EXP_COMPANY_WISE",
|
100 |
+
"86": "I-HIGH_PER",
|
101 |
+
"87": "I-INTER_PER"
|
102 |
+
},
|
103 |
+
"initializer_range": 0.02,
|
104 |
+
"intermediate_size": 3072,
|
105 |
+
"label2id": {
|
106 |
+
"B-CANDIDATE_ADDRESS": 6,
|
107 |
+
"B-CERTIFICATE": 27,
|
108 |
+
"B-C_DESIG": 3,
|
109 |
+
"B-DOB": 48,
|
110 |
+
"B-EMAIL": 52,
|
111 |
+
"B-EXP_COMPANY_WISE": 17,
|
112 |
+
"B-GRAD_PER": 46,
|
113 |
+
"B-G_CLG": 33,
|
114 |
+
"B-G_IN": 30,
|
115 |
+
"B-G_YEAR": 72,
|
116 |
+
"B-HIGH_FROM": 38,
|
117 |
+
"B-HIGH_PER": 40,
|
118 |
+
"B-INTER_FROM": 42,
|
119 |
+
"B-INTER_PER": 44,
|
120 |
+
"B-NAME": 1,
|
121 |
+
"B-PG_FROM": 58,
|
122 |
+
"B-PG_IN": 60,
|
123 |
+
"B-PG_YEAR": 69,
|
124 |
+
"B-PHONE": 9,
|
125 |
+
"B-PROJECTS": 20,
|
126 |
+
"B-SKILLS": 23,
|
127 |
+
"B-T_EXP": 14,
|
128 |
+
"I-CANDIDATE_ADDRESS": 7,
|
129 |
+
"I-CERTIFICATE": 28,
|
130 |
+
"I-C_DESIG": 4,
|
131 |
+
"I-DOB": 49,
|
132 |
+
"I-EMAIL": 53,
|
133 |
+
"I-EXP_COMPANY_WISE": 18,
|
134 |
+
"I-GRAD_PER": 82,
|
135 |
+
"I-G_CLG": 34,
|
136 |
+
"I-G_IN": 31,
|
137 |
+
"I-G_YEAR": 73,
|
138 |
+
"I-HIGH_FROM": 56,
|
139 |
+
"I-HIGH_PER": 86,
|
140 |
+
"I-INTER_FROM": 55,
|
141 |
+
"I-INTER_PER": 87,
|
142 |
+
"I-NAME": 57,
|
143 |
+
"I-PG_FROM": 65,
|
144 |
+
"I-PG_IN": 61,
|
145 |
+
"I-PG_YEAR": 70,
|
146 |
+
"I-PHONE": 10,
|
147 |
+
"I-PROJECTS": 21,
|
148 |
+
"I-SKILLS": 26,
|
149 |
+
"I-T_EXP": 15,
|
150 |
+
"L-CANDIDATE_ADDRESS": 8,
|
151 |
+
"L-CERTIFICATE": 29,
|
152 |
+
"L-C_DESIG": 5,
|
153 |
+
"L-DOB": 50,
|
154 |
+
"L-EMAIL": 54,
|
155 |
+
"L-EXP_COMPANY_WISE": 19,
|
156 |
+
"L-GRAD_PER": 47,
|
157 |
+
"L-G_CLG": 35,
|
158 |
+
"L-G_IN": 32,
|
159 |
+
"L-G_YEAR": 74,
|
160 |
+
"L-HIGH_FROM": 39,
|
161 |
+
"L-HIGH_PER": 41,
|
162 |
+
"L-INTER_FROM": 43,
|
163 |
+
"L-INTER_PER": 45,
|
164 |
+
"L-NAME": 2,
|
165 |
+
"L-PG_FROM": 59,
|
166 |
+
"L-PG_IN": 62,
|
167 |
+
"L-PG_YEAR": 71,
|
168 |
+
"L-PHONE": 11,
|
169 |
+
"L-PROJECTS": 22,
|
170 |
+
"L-SKILLS": 24,
|
171 |
+
"L-T_EXP": 16,
|
172 |
+
"O": 0,
|
173 |
+
"U-CANDIDATE_ADDRESS": 66,
|
174 |
+
"U-CERTIFICATE": 84,
|
175 |
+
"U-DOB": 13,
|
176 |
+
"U-EMAIL": 12,
|
177 |
+
"U-EXP_COMPANY_WISE": 85,
|
178 |
+
"U-GRAD_PER": 78,
|
179 |
+
"U-G_CLG": 67,
|
180 |
+
"U-G_IN": 64,
|
181 |
+
"U-G_YEAR": 36,
|
182 |
+
"U-HIGH_FROM": 77,
|
183 |
+
"U-HIGH_PER": 80,
|
184 |
+
"U-INTER_FROM": 76,
|
185 |
+
"U-INTER_PER": 79,
|
186 |
+
"U-NAME": 51,
|
187 |
+
"U-PG_FROM": 75,
|
188 |
+
"U-PG_IN": 68,
|
189 |
+
"U-PG_YEAR": 63,
|
190 |
+
"U-PHONE": 37,
|
191 |
+
"U-PROJECTS": 83,
|
192 |
+
"U-SKILLS": 25,
|
193 |
+
"U-T_EXP": 81
|
194 |
+
},
|
195 |
+
"layer_norm_eps": 1e-05,
|
196 |
+
"max_position_embeddings": 514,
|
197 |
+
"model_type": "roberta",
|
198 |
+
"num_attention_heads": 12,
|
199 |
+
"num_hidden_layers": 12,
|
200 |
+
"pad_token_id": 1,
|
201 |
+
"position_embedding_type": "absolute",
|
202 |
+
"torch_dtype": "float32",
|
203 |
+
"transformers_version": "4.35.2",
|
204 |
+
"type_vocab_size": 1,
|
205 |
+
"use_cache": true,
|
206 |
+
"vocab_size": 50265
|
207 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
special_tokens_map.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"cls_token": "<s>",
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"mask_token": {
|
6 |
+
"content": "<mask>",
|
7 |
+
"lstrip": true,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false
|
11 |
+
},
|
12 |
+
"pad_token": "<pad>",
|
13 |
+
"sep_token": "</s>",
|
14 |
+
"unk_token": "<unk>"
|
15 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"0": {
|
5 |
+
"content": "<s>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": false,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": true
|
11 |
+
},
|
12 |
+
"1": {
|
13 |
+
"content": "<pad>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": true
|
19 |
+
},
|
20 |
+
"2": {
|
21 |
+
"content": "</s>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": false,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false,
|
26 |
+
"special": true
|
27 |
+
},
|
28 |
+
"3": {
|
29 |
+
"content": "<unk>",
|
30 |
+
"lstrip": false,
|
31 |
+
"normalized": false,
|
32 |
+
"rstrip": false,
|
33 |
+
"single_word": false,
|
34 |
+
"special": true
|
35 |
+
},
|
36 |
+
"50264": {
|
37 |
+
"content": "<mask>",
|
38 |
+
"lstrip": true,
|
39 |
+
"normalized": false,
|
40 |
+
"rstrip": false,
|
41 |
+
"single_word": false,
|
42 |
+
"special": true
|
43 |
+
}
|
44 |
+
},
|
45 |
+
"bos_token": "<s>",
|
46 |
+
"clean_up_tokenization_spaces": true,
|
47 |
+
"cls_token": "<s>",
|
48 |
+
"do_lower_case": false,
|
49 |
+
"eos_token": "</s>",
|
50 |
+
"errors": "replace",
|
51 |
+
"ignore_mismatched_sizes": true,
|
52 |
+
"mask_token": "<mask>",
|
53 |
+
"model_max_length": 512,
|
54 |
+
"pad_token": "<pad>",
|
55 |
+
"sep_token": "</s>",
|
56 |
+
"tokenizer_class": "RobertaTokenizer",
|
57 |
+
"trim_offsets": true,
|
58 |
+
"unk_token": "<unk>"
|
59 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|