Hugging Face's logo

eyildirimdev
/
gliner-pii-v1

PyTorch
Model card Files
xet
 Community
eyildirimdev commited on
Commit
9bb6bdc
·
verified ·
1 Parent(s): cf361de

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. gliner_config.json +128 -0
  2. pytorch_model.bin +3 -0
  3. tokenizer.json +0 -0
  4. tokenizer_config.json +19 -0
gliner_config.json ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "class_token_index": 128001,
3
+ "decoder_mode": null,
4
+ "dropout": 0.3,
5
+ "embed_ent_token": true,
6
+ "encoder_config": {
7
+ "_name_or_path": "microsoft/deberta-v3-small",
8
+ "add_cross_attention": false,
9
+ "architectures": null,
10
+ "attention_probs_dropout_prob": 0.1,
11
+ "bos_token_id": null,
12
+ "chunk_size_feed_forward": 0,
13
+ "cross_attention_hidden_size": null,
14
+ "decoder_start_token_id": null,
15
+ "dtype": null,
16
+ "eos_token_id": null,
17
+ "finetuning_task": null,
18
+ "hidden_act": "gelu",
19
+ "hidden_dropout_prob": 0.1,
20
+ "hidden_size": 768,
21
+ "id2label": {
22
+ "0": "LABEL_0",
23
+ "1": "LABEL_1"
24
+ },
25
+ "initializer_range": 0.02,
26
+ "intermediate_size": 3072,
27
+ "is_decoder": false,
28
+ "is_encoder_decoder": false,
29
+ "label2id": {
30
+ "LABEL_0": 0,
31
+ "LABEL_1": 1
32
+ },
33
+ "layer_norm_eps": 1e-07,
34
+ "legacy": true,
35
+ "max_position_embeddings": 512,
36
+ "max_relative_positions": -1,
37
+ "model_type": "deberta-v2",
38
+ "norm_rel_ebd": "layer_norm",
39
+ "num_attention_heads": 12,
40
+ "num_hidden_layers": 6,
41
+ "output_attentions": false,
42
+ "output_hidden_states": false,
43
+ "pad_token_id": 0,
44
+ "pooler_dropout": 0,
45
+ "pooler_hidden_act": "gelu",
46
+ "pooler_hidden_size": 768,
47
+ "pos_att_type": [
48
+ "p2c",
49
+ "c2p"
50
+ ],
51
+ "position_biased_input": false,
52
+ "position_buckets": 256,
53
+ "prefix": null,
54
+ "problem_type": null,
55
+ "pruned_heads": {},
56
+ "relative_attention": true,
57
+ "return_dict": true,
58
+ "sep_token_id": null,
59
+ "share_att_key": true,
60
+ "task_specific_params": null,
61
+ "tf_legacy_loss": false,
62
+ "tie_encoder_decoder": false,
63
+ "tie_word_embeddings": true,
64
+ "tokenizer_class": null,
65
+ "torchscript": false,
66
+ "type_vocab_size": 0,
67
+ "use_bfloat16": false,
68
+ "vocab_size": 128003
69
+ },
70
+ "ent_token": "<<ENT>>",
71
+ "eval_every": 5000,
72
+ "fine_tune": true,
73
+ "freeze_token_rep": false,
74
+ "full_decoder_context": true,
75
+ "fuse_layers": false,
76
+ "has_rnn": true,
77
+ "hidden_size": 768,
78
+ "label_smoothing": 0,
79
+ "labels_decoder": null,
80
+ "labels_decoder_config": null,
81
+ "labels_encoder": null,
82
+ "labels_encoder_config": null,
83
+ "log_dir": "models/",
84
+ "loss_alpha": 0.75,
85
+ "loss_gamma": 0,
86
+ "loss_prob_margin": 0,
87
+ "loss_reduction": "sum",
88
+ "lr_encoder": "1e-5",
89
+ "lr_others": "3e-5",
90
+ "masking": "none",
91
+ "max_grad_norm": 10.0,
92
+ "max_len": 2048,
93
+ "max_neg_type_ratio": 1,
94
+ "max_types": 100,
95
+ "max_width": 12,
96
+ "model_name": "microsoft/deberta-v3-small",
97
+ "model_type": null,
98
+ "name": "span level gliner",
99
+ "neg_spans_ratio": 1.0,
100
+ "negatives": 1.0,
101
+ "num_post_fusion_layers": 1,
102
+ "num_rnn_layers": 1,
103
+ "num_steps": 30000,
104
+ "post_fusion_schema": null,
105
+ "prev_path": null,
106
+ "random_drop": true,
107
+ "represent_spans": false,
108
+ "root_dir": "gliner_logs",
109
+ "save_total_limit": 3,
110
+ "scheduler_type": "cosine",
111
+ "sep_token": "<<SEP>>",
112
+ "shuffle_types": true,
113
+ "size_sup": -1,
114
+ "span_loss_coef": 1.0,
115
+ "span_mode": "markerV0",
116
+ "subtoken_pooling": "first",
117
+ "token_loss_coef": 1.0,
118
+ "train_batch_size": 8,
119
+ "train_data": "data/full_pii.json",
120
+ "transformers_version": "5.1.0",
121
+ "use_cache": false,
122
+ "val_data_dir": "none",
123
+ "vocab_size": 128003,
124
+ "warmup_ratio": 0.05,
125
+ "weight_decay_encoder": 0.1,
126
+ "weight_decay_other": 0.01,
127
+ "words_splitter_type": "whitespace"
128
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97e3cf48f0392c20318cf44f7ad74f0604b5217787cdb8924e30271f43e26a85
3
+ size 664140735
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "clean_up_tokenization_spaces": false,
6
+ "cls_token": "[CLS]",
7
+ "do_lower_case": false,
8
+ "eos_token": "[SEP]",
9
+ "is_local": true,
10
+ "mask_token": "[MASK]",
11
+ "model_max_length": 1000000000000000019884624838656,
12
+ "pad_token": "[PAD]",
13
+ "sep_token": "[SEP]",
14
+ "sp_model_kwargs": {},
15
+ "split_by_punct": false,
16
+ "tokenizer_class": "TokenizersBackend",
17
+ "unk_token": "[UNK]",
18
+ "vocab_type": "spm"
19
+ }