Training in progress, epoch 1
Browse files- .gitignore +1 -0
- bpe.codes +0 -0
- config.json +76 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +9 -0
- tokenizer.json +0 -0
- tokenizer_config.json +13 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
bpe.codes
ADDED
The diff for this file is too large to render.
See raw diff
|
|
config.json
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "vinai/phobert-base",
|
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 |
+
"finetuning_task": "ner",
|
11 |
+
"gradient_checkpointing": false,
|
12 |
+
"hidden_act": "gelu",
|
13 |
+
"hidden_dropout_prob": 0.1,
|
14 |
+
"hidden_size": 768,
|
15 |
+
"id2label": {
|
16 |
+
"0": "O",
|
17 |
+
"1": "B-BATTERY",
|
18 |
+
"2": "I-BATTERY",
|
19 |
+
"3": "B-CAMERA",
|
20 |
+
"4": "I-CAMERA",
|
21 |
+
"5": "B-GENERAL",
|
22 |
+
"6": "I-GENERAL",
|
23 |
+
"7": "B-SER&ACC",
|
24 |
+
"8": "I-SER&ACC",
|
25 |
+
"9": "B-PERFORMANCE",
|
26 |
+
"10": "I-PERFORMANCE",
|
27 |
+
"11": "B-DESIGN",
|
28 |
+
"12": "I-DESIGN",
|
29 |
+
"13": "B-FEATURES",
|
30 |
+
"14": "I-FEATURES",
|
31 |
+
"15": "B-PRICE",
|
32 |
+
"16": "I-PRICE",
|
33 |
+
"17": "B-SCREEN",
|
34 |
+
"18": "I-SCREEN",
|
35 |
+
"19": "B-STORAGE",
|
36 |
+
"20": "I-STORAGE"
|
37 |
+
},
|
38 |
+
"initializer_range": 0.02,
|
39 |
+
"intermediate_size": 3072,
|
40 |
+
"label2id": {
|
41 |
+
"B-BATTERY": 1,
|
42 |
+
"B-CAMERA": 3,
|
43 |
+
"B-DESIGN": 11,
|
44 |
+
"B-FEATURES": 13,
|
45 |
+
"B-GENERAL": 5,
|
46 |
+
"B-PERFORMANCE": 9,
|
47 |
+
"B-PRICE": 15,
|
48 |
+
"B-SCREEN": 17,
|
49 |
+
"B-SER&ACC": 7,
|
50 |
+
"B-STORAGE": 19,
|
51 |
+
"I-BATTERY": 2,
|
52 |
+
"I-CAMERA": 4,
|
53 |
+
"I-DESIGN": 12,
|
54 |
+
"I-FEATURES": 14,
|
55 |
+
"I-GENERAL": 6,
|
56 |
+
"I-PERFORMANCE": 10,
|
57 |
+
"I-PRICE": 16,
|
58 |
+
"I-SCREEN": 18,
|
59 |
+
"I-SER&ACC": 8,
|
60 |
+
"I-STORAGE": 20,
|
61 |
+
"O": 0
|
62 |
+
},
|
63 |
+
"layer_norm_eps": 1e-05,
|
64 |
+
"max_position_embeddings": 258,
|
65 |
+
"model_type": "roberta",
|
66 |
+
"num_attention_heads": 12,
|
67 |
+
"num_hidden_layers": 12,
|
68 |
+
"pad_token_id": 1,
|
69 |
+
"position_embedding_type": "absolute",
|
70 |
+
"tokenizer_class": "PhobertTokenizer",
|
71 |
+
"torch_dtype": "float32",
|
72 |
+
"transformers_version": "4.32.0.dev0",
|
73 |
+
"type_vocab_size": 1,
|
74 |
+
"use_cache": true,
|
75 |
+
"vocab_size": 64001
|
76 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f6feb435b64a4d3df1e1b9dab3b84107b7b10ec4ea53c274d8f3a7243c397231
|
3 |
+
size 537764070
|
special_tokens_map.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"cls_token": "<s>",
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"mask_token": "<mask>",
|
6 |
+
"pad_token": "<pad>",
|
7 |
+
"sep_token": "</s>",
|
8 |
+
"unk_token": "<unk>"
|
9 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": true,
|
3 |
+
"bos_token": "<s>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"cls_token": "<s>",
|
6 |
+
"eos_token": "</s>",
|
7 |
+
"mask_token": "<mask>",
|
8 |
+
"model_max_length": 256,
|
9 |
+
"pad_token": "<pad>",
|
10 |
+
"sep_token": "</s>",
|
11 |
+
"tokenizer_class": "PhobertTokenizer",
|
12 |
+
"unk_token": "<unk>"
|
13 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e46984ad12100e80ae747f666636e9b1b60210353034c29093c088ef64e32e86
|
3 |
+
size 4408
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|