ClauseGuard CUAD classifier — ONNX + dynamic INT8 quantization (Legal-BERT + LoRA merged)
Browse files- config.json +113 -0
- model_quantized.onnx +3 -0
- ort_config.json +33 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_ids": 0,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "Document Name",
|
| 15 |
+
"1": "Parties",
|
| 16 |
+
"2": "Agreement Date",
|
| 17 |
+
"3": "Effective Date",
|
| 18 |
+
"4": "Expiration Date",
|
| 19 |
+
"5": "Renewal Term",
|
| 20 |
+
"6": "Notice Period to Terminate Renewal",
|
| 21 |
+
"7": "Governing Law",
|
| 22 |
+
"8": "Most Favored Nation",
|
| 23 |
+
"9": "Non-Compete",
|
| 24 |
+
"10": "Exclusivity",
|
| 25 |
+
"11": "No-Solicit of Customers",
|
| 26 |
+
"12": "No-Solicit of Employees",
|
| 27 |
+
"13": "Non-Disparagement",
|
| 28 |
+
"14": "Termination for Convenience",
|
| 29 |
+
"15": "ROFR/ROFO/ROFN",
|
| 30 |
+
"16": "Change of Control",
|
| 31 |
+
"17": "Anti-Assignment",
|
| 32 |
+
"18": "Revenue/Profit Sharing",
|
| 33 |
+
"19": "Price Restriction",
|
| 34 |
+
"20": "Minimum Commitment",
|
| 35 |
+
"21": "Volume Restriction",
|
| 36 |
+
"22": "IP Ownership Assignment",
|
| 37 |
+
"23": "Joint IP Ownership",
|
| 38 |
+
"24": "License Grant",
|
| 39 |
+
"25": "Non-Transferable License",
|
| 40 |
+
"26": "Affiliate License-Licensor",
|
| 41 |
+
"27": "Affiliate License-Licensee",
|
| 42 |
+
"28": "Unlimited/All-You-Can-Eat License",
|
| 43 |
+
"29": "Irrevocable or Perpetual License",
|
| 44 |
+
"30": "Source Code Escrow",
|
| 45 |
+
"31": "Post-Termination Services",
|
| 46 |
+
"32": "Audit Rights",
|
| 47 |
+
"33": "Uncapped Liability",
|
| 48 |
+
"34": "Cap on Liability",
|
| 49 |
+
"35": "Liquidated Damages",
|
| 50 |
+
"36": "Warranty Duration",
|
| 51 |
+
"37": "Insurance",
|
| 52 |
+
"38": "Covenant Not to Sue",
|
| 53 |
+
"39": "Third Party Beneficiary",
|
| 54 |
+
"40": "Other"
|
| 55 |
+
},
|
| 56 |
+
"initializer_range": 0.02,
|
| 57 |
+
"intermediate_size": 3072,
|
| 58 |
+
"label2id": {
|
| 59 |
+
"Affiliate License-Licensee": 27,
|
| 60 |
+
"Affiliate License-Licensor": 26,
|
| 61 |
+
"Agreement Date": 2,
|
| 62 |
+
"Anti-Assignment": 17,
|
| 63 |
+
"Audit Rights": 32,
|
| 64 |
+
"Cap on Liability": 34,
|
| 65 |
+
"Change of Control": 16,
|
| 66 |
+
"Covenant Not to Sue": 38,
|
| 67 |
+
"Document Name": 0,
|
| 68 |
+
"Effective Date": 3,
|
| 69 |
+
"Exclusivity": 10,
|
| 70 |
+
"Expiration Date": 4,
|
| 71 |
+
"Governing Law": 7,
|
| 72 |
+
"IP Ownership Assignment": 22,
|
| 73 |
+
"Insurance": 37,
|
| 74 |
+
"Irrevocable or Perpetual License": 29,
|
| 75 |
+
"Joint IP Ownership": 23,
|
| 76 |
+
"License Grant": 24,
|
| 77 |
+
"Liquidated Damages": 35,
|
| 78 |
+
"Minimum Commitment": 20,
|
| 79 |
+
"Most Favored Nation": 8,
|
| 80 |
+
"No-Solicit of Customers": 11,
|
| 81 |
+
"No-Solicit of Employees": 12,
|
| 82 |
+
"Non-Compete": 9,
|
| 83 |
+
"Non-Disparagement": 13,
|
| 84 |
+
"Non-Transferable License": 25,
|
| 85 |
+
"Notice Period to Terminate Renewal": 6,
|
| 86 |
+
"Other": 40,
|
| 87 |
+
"Parties": 1,
|
| 88 |
+
"Post-Termination Services": 31,
|
| 89 |
+
"Price Restriction": 19,
|
| 90 |
+
"ROFR/ROFO/ROFN": 15,
|
| 91 |
+
"Renewal Term": 5,
|
| 92 |
+
"Revenue/Profit Sharing": 18,
|
| 93 |
+
"Source Code Escrow": 30,
|
| 94 |
+
"Termination for Convenience": 14,
|
| 95 |
+
"Third Party Beneficiary": 39,
|
| 96 |
+
"Uncapped Liability": 33,
|
| 97 |
+
"Unlimited/All-You-Can-Eat License": 28,
|
| 98 |
+
"Volume Restriction": 21,
|
| 99 |
+
"Warranty Duration": 36
|
| 100 |
+
},
|
| 101 |
+
"layer_norm_eps": 1e-12,
|
| 102 |
+
"max_position_embeddings": 512,
|
| 103 |
+
"model_type": "bert",
|
| 104 |
+
"num_attention_heads": 12,
|
| 105 |
+
"num_hidden_layers": 12,
|
| 106 |
+
"output_past": true,
|
| 107 |
+
"pad_token_id": 0,
|
| 108 |
+
"position_embedding_type": "absolute",
|
| 109 |
+
"transformers_version": "4.57.6",
|
| 110 |
+
"type_vocab_size": 2,
|
| 111 |
+
"use_cache": true,
|
| 112 |
+
"vocab_size": 30522
|
| 113 |
+
}
|
model_quantized.onnx
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd34a472056553ac06cf7538107720a29f3a1b6ceea3e34921f6022c10fef664
|
| 3 |
+
size 110310911
|
ort_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"one_external_file": true,
|
| 3 |
+
"opset": null,
|
| 4 |
+
"optimization": {},
|
| 5 |
+
"quantization": {
|
| 6 |
+
"activations_dtype": "QUInt8",
|
| 7 |
+
"activations_symmetric": false,
|
| 8 |
+
"format": "QOperator",
|
| 9 |
+
"is_static": false,
|
| 10 |
+
"mode": "IntegerOps",
|
| 11 |
+
"nodes_to_exclude": [],
|
| 12 |
+
"nodes_to_quantize": [],
|
| 13 |
+
"operators_to_quantize": [
|
| 14 |
+
"Conv",
|
| 15 |
+
"MatMul",
|
| 16 |
+
"Attention",
|
| 17 |
+
"LSTM",
|
| 18 |
+
"Gather",
|
| 19 |
+
"Transpose",
|
| 20 |
+
"EmbedLayerNormalization"
|
| 21 |
+
],
|
| 22 |
+
"per_channel": false,
|
| 23 |
+
"qdq_add_pair_to_weight": false,
|
| 24 |
+
"qdq_dedicated_pair": false,
|
| 25 |
+
"qdq_op_type_per_channel_support_to_axis": {
|
| 26 |
+
"MatMul": 1
|
| 27 |
+
},
|
| 28 |
+
"reduce_range": false,
|
| 29 |
+
"weights_dtype": "QInt8",
|
| 30 |
+
"weights_symmetric": true
|
| 31 |
+
},
|
| 32 |
+
"use_external_data_format": false
|
| 33 |
+
}
|
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 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|