Commit
·
31c3149
1
Parent(s):
29f2cf0
Training in progress, epoch 1
Browse files- .gitignore +1 -0
- config.json +119 -0
- pytorch_model.bin +3 -0
- runs/Mar20_08-26-22_t1-45-bhs5/1679300789.8015804/events.out.tfevents.1679300789.t1-45-bhs5.1856.1 +3 -0
- runs/Mar20_08-26-22_t1-45-bhs5/events.out.tfevents.1679300789.t1-45-bhs5.1856.0 +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,119 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "distilbert-base-multilingual-cased",
|
3 |
+
"activation": "gelu",
|
4 |
+
"architectures": [
|
5 |
+
"DistilBertForTokenClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"dim": 768,
|
9 |
+
"dropout": 0.1,
|
10 |
+
"hidden_dim": 3072,
|
11 |
+
"id2label": {
|
12 |
+
"0": "O",
|
13 |
+
"1": "B-BUILDING",
|
14 |
+
"2": "I-BUILDING",
|
15 |
+
"3": "B-CITY",
|
16 |
+
"4": "I-CITY",
|
17 |
+
"5": "B-COUNTRY",
|
18 |
+
"6": "I-COUNTRY",
|
19 |
+
"7": "B-PLACE",
|
20 |
+
"8": "I-PLACE",
|
21 |
+
"9": "B-TERRITORY",
|
22 |
+
"10": "I-TERRITORY",
|
23 |
+
"11": "I-UNIT",
|
24 |
+
"12": "B-UNIT",
|
25 |
+
"13": "B-VALUE",
|
26 |
+
"14": "I-VALUE",
|
27 |
+
"15": "B-YEAR",
|
28 |
+
"16": "I-YEAR",
|
29 |
+
"17": "B-STANDARD ABBREVIATION",
|
30 |
+
"18": "I-STANDARD ABBREVIATION",
|
31 |
+
"19": "B-MONTH",
|
32 |
+
"20": "I-MONTH",
|
33 |
+
"21": "B-DAY",
|
34 |
+
"22": "I-DAY",
|
35 |
+
"23": "B-AGE",
|
36 |
+
"24": "I-AGE",
|
37 |
+
"25": "B-ETHNIC CATEGORY",
|
38 |
+
"26": "I-ETHNIC CATEGORY",
|
39 |
+
"27": "B-FAMILY NAME",
|
40 |
+
"28": "I-FAMILY NAME",
|
41 |
+
"29": "B-INITIAL NAME",
|
42 |
+
"30": "I-INITIAL NAME",
|
43 |
+
"31": "B-MARITAL STATUS",
|
44 |
+
"32": "I-MARITAL STATUS",
|
45 |
+
"33": "B-PROFESSION",
|
46 |
+
"34": "I-PROFESSION",
|
47 |
+
"35": "B-ROLE",
|
48 |
+
"36": "I-ROLE",
|
49 |
+
"37": "B-NATIONALITY",
|
50 |
+
"38": "I-NATIONALITY",
|
51 |
+
"39": "B-TITLE",
|
52 |
+
"40": "I-TITLE",
|
53 |
+
"41": "B-URL",
|
54 |
+
"42": "I-URL",
|
55 |
+
"43": "B-TYPE",
|
56 |
+
"44": "I-TYPE"
|
57 |
+
},
|
58 |
+
"initializer_range": 0.02,
|
59 |
+
"label2id": {
|
60 |
+
"B-AGE": 23,
|
61 |
+
"B-BUILDING": 1,
|
62 |
+
"B-CITY": 3,
|
63 |
+
"B-COUNTRY": 5,
|
64 |
+
"B-DAY": 21,
|
65 |
+
"B-ETHNIC CATEGORY": 25,
|
66 |
+
"B-FAMILY NAME": 27,
|
67 |
+
"B-INITIAL NAME": 29,
|
68 |
+
"B-MARITAL STATUS": 31,
|
69 |
+
"B-MONTH": 19,
|
70 |
+
"B-NATIONALITY": 37,
|
71 |
+
"B-PLACE": 7,
|
72 |
+
"B-PROFESSION": 33,
|
73 |
+
"B-ROLE": 35,
|
74 |
+
"B-STANDARD ABBREVIATION": 17,
|
75 |
+
"B-TERRITORY": 9,
|
76 |
+
"B-TITLE": 39,
|
77 |
+
"B-TYPE": 43,
|
78 |
+
"B-UNIT": 12,
|
79 |
+
"B-URL": 41,
|
80 |
+
"B-VALUE": 13,
|
81 |
+
"B-YEAR": 15,
|
82 |
+
"I-AGE": 24,
|
83 |
+
"I-BUILDING": 2,
|
84 |
+
"I-CITY": 4,
|
85 |
+
"I-COUNTRY": 6,
|
86 |
+
"I-DAY": 22,
|
87 |
+
"I-ETHNIC CATEGORY": 26,
|
88 |
+
"I-FAMILY NAME": 28,
|
89 |
+
"I-INITIAL NAME": 30,
|
90 |
+
"I-MARITAL STATUS": 32,
|
91 |
+
"I-MONTH": 20,
|
92 |
+
"I-NATIONALITY": 38,
|
93 |
+
"I-PLACE": 8,
|
94 |
+
"I-PROFESSION": 34,
|
95 |
+
"I-ROLE": 36,
|
96 |
+
"I-STANDARD ABBREVIATION": 18,
|
97 |
+
"I-TERRITORY": 10,
|
98 |
+
"I-TITLE": 40,
|
99 |
+
"I-TYPE": 44,
|
100 |
+
"I-UNIT": 11,
|
101 |
+
"I-URL": 42,
|
102 |
+
"I-VALUE": 14,
|
103 |
+
"I-YEAR": 16,
|
104 |
+
"O": 0
|
105 |
+
},
|
106 |
+
"max_position_embeddings": 512,
|
107 |
+
"model_type": "distilbert",
|
108 |
+
"n_heads": 12,
|
109 |
+
"n_layers": 6,
|
110 |
+
"output_past": true,
|
111 |
+
"pad_token_id": 0,
|
112 |
+
"qa_dropout": 0.1,
|
113 |
+
"seq_classif_dropout": 0.2,
|
114 |
+
"sinusoidal_pos_embds": false,
|
115 |
+
"tie_weights_": true,
|
116 |
+
"torch_dtype": "float32",
|
117 |
+
"transformers_version": "4.26.0",
|
118 |
+
"vocab_size": 119547
|
119 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8b0f729ad5a54f2a513fe2a996c4bceefc45ad0ae7a3a86f545d031b6cdda22d
|
3 |
+
size 539109541
|
runs/Mar20_08-26-22_t1-45-bhs5/1679300789.8015804/events.out.tfevents.1679300789.t1-45-bhs5.1856.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f1c4462a6baf3e9c88899075a53d67e694bbbc12d7e076584e0745f1cd3cbd0e
|
3 |
+
size 5758
|
runs/Mar20_08-26-22_t1-45-bhs5/events.out.tfevents.1679300789.t1-45-bhs5.1856.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:692d9533d3433aa23e2236bd0de7b46e72cecb108cc0ba0639fff091e0d3b5e5
|
3 |
+
size 6801
|
special_tokens_map.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"mask_token": "[MASK]",
|
4 |
+
"pad_token": "[PAD]",
|
5 |
+
"sep_token": "[SEP]",
|
6 |
+
"unk_token": "[UNK]"
|
7 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cls_token": "[CLS]",
|
3 |
+
"do_lower_case": false,
|
4 |
+
"mask_token": "[MASK]",
|
5 |
+
"model_max_length": 512,
|
6 |
+
"name_or_path": "distilbert-base-multilingual-cased",
|
7 |
+
"pad_token": "[PAD]",
|
8 |
+
"sep_token": "[SEP]",
|
9 |
+
"special_tokens_map_file": null,
|
10 |
+
"strip_accents": null,
|
11 |
+
"tokenize_chinese_chars": true,
|
12 |
+
"tokenizer_class": "DistilBertTokenizer",
|
13 |
+
"unk_token": "[UNK]"
|
14 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7f8d8bd2154bf0ef5d7ba90b90854c09b8b27c76986e47208130be7fb18419ad
|
3 |
+
size 3579
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|