amicus-ner-v2: LoRA fine-tuned on Nigerian legal corpus (regex silver labels + Gemini augmentation)
Browse files- config.json +69 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +22 -0
config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_cross_attention": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": null,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"eos_token_id": null,
|
| 11 |
+
"eos_token_ids": 0,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.1,
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "O",
|
| 17 |
+
"1": "B-CASE_NAME",
|
| 18 |
+
"2": "I-CASE_NAME",
|
| 19 |
+
"3": "B-CITATION",
|
| 20 |
+
"4": "I-CITATION",
|
| 21 |
+
"5": "B-STATUTE",
|
| 22 |
+
"6": "I-STATUTE",
|
| 23 |
+
"7": "B-COURT",
|
| 24 |
+
"8": "I-COURT",
|
| 25 |
+
"9": "B-DATE",
|
| 26 |
+
"10": "I-DATE",
|
| 27 |
+
"11": "B-JUDGE",
|
| 28 |
+
"12": "I-JUDGE",
|
| 29 |
+
"13": "B-RATIO",
|
| 30 |
+
"14": "I-RATIO",
|
| 31 |
+
"15": "B-HELD",
|
| 32 |
+
"16": "I-HELD"
|
| 33 |
+
},
|
| 34 |
+
"initializer_range": 0.02,
|
| 35 |
+
"intermediate_size": 3072,
|
| 36 |
+
"is_decoder": false,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"B-CASE_NAME": 1,
|
| 39 |
+
"B-CITATION": 3,
|
| 40 |
+
"B-COURT": 7,
|
| 41 |
+
"B-DATE": 9,
|
| 42 |
+
"B-HELD": 15,
|
| 43 |
+
"B-JUDGE": 11,
|
| 44 |
+
"B-RATIO": 13,
|
| 45 |
+
"B-STATUTE": 5,
|
| 46 |
+
"I-CASE_NAME": 2,
|
| 47 |
+
"I-CITATION": 4,
|
| 48 |
+
"I-COURT": 8,
|
| 49 |
+
"I-DATE": 10,
|
| 50 |
+
"I-HELD": 16,
|
| 51 |
+
"I-JUDGE": 12,
|
| 52 |
+
"I-RATIO": 14,
|
| 53 |
+
"I-STATUTE": 6,
|
| 54 |
+
"O": 0
|
| 55 |
+
},
|
| 56 |
+
"layer_norm_eps": 1e-12,
|
| 57 |
+
"max_position_embeddings": 512,
|
| 58 |
+
"model_type": "bert",
|
| 59 |
+
"num_attention_heads": 12,
|
| 60 |
+
"num_hidden_layers": 12,
|
| 61 |
+
"output_past": true,
|
| 62 |
+
"pad_token_id": 0,
|
| 63 |
+
"position_embedding_type": "absolute",
|
| 64 |
+
"tie_word_embeddings": true,
|
| 65 |
+
"transformers_version": "5.12.1",
|
| 66 |
+
"type_vocab_size": 2,
|
| 67 |
+
"use_cache": false,
|
| 68 |
+
"vocab_size": 30522
|
| 69 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09de8adf267c16a9a35765e158e6983e90f3e9eb8b009a6fbf3729fda0fc071e
|
| 3 |
+
size 435642228
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
+
"cls_token": "[CLS]",
|
| 5 |
+
"do_basic_tokenize": true,
|
| 6 |
+
"do_lower_case": true,
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"mask_token": "[MASK]",
|
| 10 |
+
"max_length": 256,
|
| 11 |
+
"model_max_length": 512,
|
| 12 |
+
"never_split": null,
|
| 13 |
+
"pad_token": "[PAD]",
|
| 14 |
+
"sep_token": "[SEP]",
|
| 15 |
+
"stride": 0,
|
| 16 |
+
"strip_accents": null,
|
| 17 |
+
"tokenize_chinese_chars": true,
|
| 18 |
+
"tokenizer_class": "BertTokenizer",
|
| 19 |
+
"truncation_side": "right",
|
| 20 |
+
"truncation_strategy": "longest_first",
|
| 21 |
+
"unk_token": "[UNK]"
|
| 22 |
+
}
|