Hugging Face's logo

khalidrajan
/
roberta-base_legal_ner_finetuned

TensorBoard
Safetensors
roberta
Generated from Trainer
Model card Files Metrics Community
khalidrajan commited on
Commit
f5ec09e
•
1 Parent(s): 77bcbdc

End of training

Browse files
Files changed (9) hide show
  1. README.md +26 -26
  2. config.json +14 -14
  3. merges.txt +0 -0
  4. model.safetensors +1 -1
  5. special_tokens_map.json +15 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +57 -0
  8. training_args.bin +1 -1
  9. vocab.json +0 -0
README.md CHANGED
@@ -15,27 +15,27 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 0.2446
19
- - Law Precision: 0.8692
20
  - Law Recall: 0.8692
21
- - Law F1: 0.8692
22
  - Law Number: 107
23
- - Violated by Precision: 0.8710
24
- - Violated by Recall: 0.7606
25
- - Violated by F1: 0.8120
26
  - Violated by Number: 71
27
- - Violated on Precision: 0.4638
28
- - Violated on Recall: 0.5
29
- - Violated on F1: 0.4812
30
  - Violated on Number: 64
31
- - Violation Precision: 0.6539
32
- - Violation Recall: 0.6872
33
- - Violation F1: 0.6701
34
  - Violation Number: 374
35
- - Overall Precision: 0.6910
36
- - Overall Recall: 0.7078
37
- - Overall F1: 0.6993
38
- - Overall Accuracy: 0.9536
39
 
40
  ## Model description
41
 
@@ -67,16 +67,16 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Law Precision | Law Recall | Law F1 | Law Number | Violated by Precision | Violated by Recall | Violated by F1 | Violated by Number | Violated on Precision | Violated on Recall | Violated on F1 | Violated on Number | Violation Precision | Violation Recall | Violation F1 | Violation Number | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:-------------:|:----------:|:------:|:----------:|:---------------------:|:------------------:|:--------------:|:------------------:|:---------------------:|:------------------:|:--------------:|:------------------:|:-------------------:|:----------------:|:------------:|:----------------:|:-----------------:|:--------------:|:----------:|:----------------:|
70
- | No log | 1.0 | 85 | 0.6877 | 0.0 | 0.0 | 0.0 | 107 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 64 | 0.0 | 0.0 | 0.0 | 374 | 0.0 | 0.0 | 0.0 | 0.7713 |
71
- | No log | 2.0 | 170 | 0.3547 | 0.0 | 0.0 | 0.0 | 107 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 64 | 0.2547 | 0.3235 | 0.2850 | 374 | 0.2537 | 0.1964 | 0.2214 | 0.8903 |
72
- | No log | 3.0 | 255 | 0.2770 | 0.38 | 0.1776 | 0.2420 | 107 | 0.0 | 0.0 | 0.0 | 71 | 0.0714 | 0.0156 | 0.0256 | 64 | 0.4029 | 0.4492 | 0.4248 | 374 | 0.3909 | 0.3052 | 0.3428 | 0.9125 |
73
- | No log | 4.0 | 340 | 0.2137 | 0.7143 | 0.7477 | 0.7306 | 107 | 0.3214 | 0.2535 | 0.2835 | 71 | 0.2323 | 0.3594 | 0.2822 | 64 | 0.4668 | 0.5267 | 0.4950 | 374 | 0.4615 | 0.5162 | 0.4874 | 0.9327 |
74
- | No log | 5.0 | 425 | 0.1779 | 0.7672 | 0.8318 | 0.7982 | 107 | 0.7097 | 0.6197 | 0.6617 | 71 | 0.3667 | 0.3438 | 0.3548 | 64 | 0.5653 | 0.6364 | 0.5987 | 374 | 0.5964 | 0.6380 | 0.6165 | 0.9462 |
75
- | 0.4985 | 6.0 | 510 | 0.2284 | 0.7876 | 0.8318 | 0.8091 | 107 | 0.6849 | 0.7042 | 0.6944 | 71 | 0.5106 | 0.375 | 0.4324 | 64 | 0.5968 | 0.7005 | 0.6445 | 374 | 0.6324 | 0.6899 | 0.6599 | 0.9459 |
76
- | 0.4985 | 7.0 | 595 | 0.2240 | 0.8214 | 0.8598 | 0.8402 | 107 | 0.7969 | 0.7183 | 0.7556 | 71 | 0.4848 | 0.5 | 0.4923 | 64 | 0.6203 | 0.6551 | 0.6372 | 374 | 0.6593 | 0.6818 | 0.6704 | 0.9483 |
77
- | 0.4985 | 8.0 | 680 | 0.1989 | 0.8407 | 0.8879 | 0.8636 | 107 | 0.7143 | 0.7746 | 0.7432 | 71 | 0.4079 | 0.4844 | 0.4429 | 64 | 0.6658 | 0.6979 | 0.6815 | 374 | 0.6717 | 0.7175 | 0.6939 | 0.9517 |
78
- | 0.4985 | 9.0 | 765 | 0.2474 | 0.8559 | 0.8879 | 0.8716 | 107 | 0.7971 | 0.7746 | 0.7857 | 71 | 0.4429 | 0.4844 | 0.4627 | 64 | 0.6539 | 0.6872 | 0.6701 | 374 | 0.6812 | 0.7110 | 0.6958 | 0.9506 |
79
- | 0.4985 | 10.0 | 850 | 0.2446 | 0.8692 | 0.8692 | 0.8692 | 107 | 0.8710 | 0.7606 | 0.8120 | 71 | 0.4638 | 0.5 | 0.4812 | 64 | 0.6539 | 0.6872 | 0.6701 | 374 | 0.6910 | 0.7078 | 0.6993 | 0.9536 |
80
 
81
 
82
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.2520
19
+ - Law Precision: 0.8774
20
  - Law Recall: 0.8692
21
+ - Law F1: 0.8732
22
  - Law Number: 107
23
+ - Violated by Precision: 0.8226
24
+ - Violated by Recall: 0.7183
25
+ - Violated by F1: 0.7669
26
  - Violated by Number: 71
27
+ - Violated on Precision: 0.5312
28
+ - Violated on Recall: 0.5312
29
+ - Violated on F1: 0.5312
30
  - Violated on Number: 64
31
+ - Violation Precision: 0.6625
32
+ - Violation Recall: 0.7032
33
+ - Violation F1: 0.6822
34
  - Violation Number: 374
35
+ - Overall Precision: 0.7011
36
+ - Overall Recall: 0.7159
37
+ - Overall F1: 0.7084
38
+ - Overall Accuracy: 0.9546
39
 
40
  ## Model description
41
 
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Law Precision | Law Recall | Law F1 | Law Number | Violated by Precision | Violated by Recall | Violated by F1 | Violated by Number | Violated on Precision | Violated on Recall | Violated on F1 | Violated on Number | Violation Precision | Violation Recall | Violation F1 | Violation Number | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
69
  |:-------------:|:-----:|:----:|:---------------:|:-------------:|:----------:|:------:|:----------:|:---------------------:|:------------------:|:--------------:|:------------------:|:---------------------:|:------------------:|:--------------:|:------------------:|:-------------------:|:----------------:|:------------:|:----------------:|:-----------------:|:--------------:|:----------:|:----------------:|
70
+ | No log | 1.0 | 85 | 0.7163 | 0.0 | 0.0 | 0.0 | 107 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 64 | 0.0 | 0.0 | 0.0 | 374 | 0.0 | 0.0 | 0.0 | 0.7706 |
71
+ | No log | 2.0 | 170 | 0.3725 | 0.0 | 0.0 | 0.0 | 107 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 64 | 0.2539 | 0.3048 | 0.2770 | 374 | 0.2539 | 0.1851 | 0.2141 | 0.8888 |
72
+ | No log | 3.0 | 255 | 0.2796 | 0.3542 | 0.1589 | 0.2194 | 107 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 64 | 0.4167 | 0.4813 | 0.4467 | 374 | 0.4045 | 0.3198 | 0.3572 | 0.9130 |
73
+ | No log | 4.0 | 340 | 0.2083 | 0.7692 | 0.7477 | 0.7583 | 107 | 0.72 | 0.5070 | 0.5950 | 71 | 0.4694 | 0.3594 | 0.4071 | 64 | 0.5340 | 0.5882 | 0.5598 | 374 | 0.5837 | 0.5828 | 0.5833 | 0.9392 |
74
+ | No log | 5.0 | 425 | 0.1879 | 0.8131 | 0.8131 | 0.8131 | 107 | 0.6984 | 0.6197 | 0.6567 | 71 | 0.5254 | 0.4844 | 0.5041 | 64 | 0.5714 | 0.6524 | 0.6092 | 374 | 0.6189 | 0.6591 | 0.6384 | 0.9441 |
75
+ | 0.5159 | 6.0 | 510 | 0.2667 | 0.7627 | 0.8411 | 0.8000 | 107 | 0.6901 | 0.6901 | 0.6901 | 71 | 0.5 | 0.375 | 0.4286 | 64 | 0.5944 | 0.6230 | 0.6084 | 374 | 0.6296 | 0.6429 | 0.6361 | 0.9382 |
76
+ | 0.5159 | 7.0 | 595 | 0.2122 | 0.8198 | 0.8505 | 0.8349 | 107 | 0.7619 | 0.6761 | 0.7164 | 71 | 0.5231 | 0.5312 | 0.5271 | 64 | 0.6171 | 0.6551 | 0.6355 | 374 | 0.6572 | 0.6786 | 0.6677 | 0.9500 |
77
+ | 0.5159 | 8.0 | 680 | 0.2445 | 0.8571 | 0.8411 | 0.8491 | 107 | 0.7143 | 0.7746 | 0.7432 | 71 | 0.5075 | 0.5312 | 0.5191 | 64 | 0.6048 | 0.6791 | 0.6398 | 374 | 0.6472 | 0.7029 | 0.6739 | 0.9494 |
78
+ | 0.5159 | 9.0 | 765 | 0.2679 | 0.8468 | 0.8785 | 0.8624 | 107 | 0.85 | 0.7183 | 0.7786 | 71 | 0.4648 | 0.5156 | 0.4889 | 64 | 0.6170 | 0.6979 | 0.6550 | 374 | 0.6602 | 0.7127 | 0.6854 | 0.9501 |
79
+ | 0.5159 | 10.0 | 850 | 0.2520 | 0.8774 | 0.8692 | 0.8732 | 107 | 0.8226 | 0.7183 | 0.7669 | 71 | 0.5312 | 0.5312 | 0.5312 | 64 | 0.6625 | 0.7032 | 0.6822 | 374 | 0.7011 | 0.7159 | 0.7084 | 0.9546 |
80
 
81
 
82
  ### Framework versions
config.json CHANGED
@@ -11,28 +11,28 @@
11
  "hidden_dropout_prob": 0.1,
12
  "hidden_size": 768,
13
  "id2label": {
14
- "0": "I-VIOLATED BY",
15
- "1": "I-VIOLATION",
16
- "2": "B-VIOLATED BY",
17
- "3": "I-VIOLATED ON",
18
- "4": "B-LAW",
19
  "5": "B-VIOLATED ON",
20
- "6": "O",
21
  "7": "I-LAW",
22
- "8": "B-VIOLATION"
23
  },
24
  "initializer_range": 0.02,
25
  "intermediate_size": 3072,
26
  "label2id": {
27
- "B-LAW": 4,
28
- "B-VIOLATED BY": 2,
29
  "B-VIOLATED ON": 5,
30
- "B-VIOLATION": 8,
31
  "I-LAW": 7,
32
- "I-VIOLATED BY": 0,
33
- "I-VIOLATED ON": 3,
34
- "I-VIOLATION": 1,
35
- "O": 6
36
  },
37
  "layer_norm_eps": 1e-05,
38
  "max_position_embeddings": 514,
 
11
  "hidden_dropout_prob": 0.1,
12
  "hidden_size": 768,
13
  "id2label": {
14
+ "0": "O",
15
+ "1": "B-VIOLATION",
16
+ "2": "I-VIOLATED ON",
17
+ "3": "B-VIOLATED BY",
18
+ "4": "I-VIOLATED BY",
19
  "5": "B-VIOLATED ON",
20
+ "6": "B-LAW",
21
  "7": "I-LAW",
22
+ "8": "I-VIOLATION"
23
  },
24
  "initializer_range": 0.02,
25
  "intermediate_size": 3072,
26
  "label2id": {
27
+ "B-LAW": 6,
28
+ "B-VIOLATED BY": 3,
29
  "B-VIOLATED ON": 5,
30
+ "B-VIOLATION": 1,
31
  "I-LAW": 7,
32
+ "I-VIOLATED BY": 4,
33
+ "I-VIOLATED ON": 2,
34
+ "I-VIOLATION": 8,
35
+ "O": 0
36
  },
37
  "layer_norm_eps": 1e-05,
38
  "max_position_embeddings": 514,
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:79b98d7a2654d2bfbd70caa94ee1d4b1e304ed46c6d921867ca7d47bbecb4278
3
  size 496271772
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2976dcfaa487844aad39a35b27dac9e936ba0e1c85baf507f7245e6afda2b235
3
  size 496271772
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 512,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "RobertaTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:968e436fa01a01cc62712dc9086c80e4cfd999f0897efa2b21ea24bdb91039f5
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05d889f9e7552e08630ec7acb7465da2bf23632530e0adeea75969c962466103
3
  size 5240
vocab.json ADDED
The diff for this file is too large to render. See raw diff