Hugging Face's logo

saattrupdan
/
employment-contract-ner-da

Token Classification
Transformers
PyTorch
TensorBoard
Safetensors
Danish
xlm-roberta
Inference Endpoints
Model card Files Metrics Community
2
saattrupdan commited on
Commit
4e03af4
•
1 Parent(s): 88a1b98

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +69 -40
README.md CHANGED
@@ -1,27 +1,33 @@
1
  ---
2
- language:
3
- - da
4
- license: mit
5
  model-index:
6
  - name: contract-ner-model-da
7
- results:
8
- - task:
9
- type: token-classification
10
- name: Token Classification
11
- widget:
12
- - "Medarbejderen ansættes til 35.000,00 kr. om måneden og arbejdsstedet er Supergaden 21, 2000 Frederiksberg."
13
- inference:
14
- parameters:
15
- aggregation_strategy: "first"
16
  ---
17
 
 
 
 
18
  # contract-ner-model-da
19
 
20
- This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on a custom contracts dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.0100
23
- - Micro F1: 0.9074
24
- - Micro F1 No Misc: 0.9074
 
 
 
 
 
 
 
 
 
 
 
25
 
26
  ## Training procedure
27
 
@@ -36,33 +42,56 @@ The following hyperparameters were used during training:
36
  - total_train_batch_size: 32
37
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
38
  - lr_scheduler_type: linear
39
- - lr_scheduler_warmup_steps: 4356
40
- - num_epochs: 100
41
 
42
  ### Training results
43
 
44
- | Training Loss | Epoch | Step | Validation Loss | Micro F1 | Micro F1 No Misc |
45
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:----------------:|
46
- | 1.1722 | 4.88 | 200 | 0.0612 | 0.0 | 0.0 |
47
- | 0.0341 | 9.76 | 400 | 0.0209 | 0.0 | 0.0 |
48
- | 0.0158 | 14.63 | 600 | 0.0109 | 0.7233 | 0.7233 |
49
- | 0.0085 | 19.51 | 800 | 0.0097 | 0.7727 | 0.7727 |
50
- | 0.0057 | 24.39 | 1000 | 0.0084 | 0.8319 | 0.8319 |
51
- | 0.0041 | 29.27 | 1200 | 0.0085 | 0.8630 | 0.8630 |
52
- | 0.0031 | 34.15 | 1400 | 0.0089 | 0.8261 | 0.8261 |
53
- | 0.0023 | 39.02 | 1600 | 0.0066 | 0.8785 | 0.8785 |
54
- | 0.0017 | 43.9 | 1800 | 0.0106 | 0.8164 | 0.8164 |
55
- | 0.0012 | 48.78 | 2000 | 0.0092 | 0.8730 | 0.8730 |
56
- | 0.0008 | 53.66 | 2200 | 0.0076 | 0.8868 | 0.8868 |
57
- | 0.0007 | 58.54 | 2400 | 0.0075 | 0.9017 | 0.9017 |
58
- | 0.0005 | 63.41 | 2600 | 0.0096 | 0.8806 | 0.8806 |
59
- | 0.0004 | 68.29 | 2800 | 0.0094 | 0.8852 | 0.8852 |
60
- | 0.0004 | 73.17 | 3000 | 0.0084 | 0.9126 | 0.9126 |
61
- | 0.0003 | 78.05 | 3200 | 0.0083 | 0.8986 | 0.8986 |
62
- | 0.0003 | 82.93 | 3400 | 0.0093 | 0.9144 | 0.9144 |
63
- | 0.0003 | 87.8 | 3600 | 0.0088 | 0.9231 | 0.9231 |
64
- | 0.0001 | 92.68 | 3800 | 0.0080 | 0.9280 | 0.9280 |
65
- | 0.0002 | 97.56 | 4000 | 0.0100 | 0.9074 | 0.9074 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
 
67
 
68
  ### Framework versions
 
1
  ---
2
+ license: cc-by-4.0
3
+ tags:
4
+ - generated_from_trainer
5
  model-index:
6
  - name: contract-ner-model-da
7
+ results: []
 
 
 
 
 
 
 
 
8
  ---
9
 
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
  # contract-ner-model-da
14
 
15
+ This model is a fine-tuned version of [NbAiLab/nb-bert-base](https://huggingface.co/NbAiLab/nb-bert-base) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.0213
18
+ - Micro F1: 0.9217
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
 
32
  ## Training procedure
33
 
 
42
  - total_train_batch_size: 32
43
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
  - lr_scheduler_type: linear
45
+ - lr_scheduler_warmup_steps: 201
46
+ - num_epochs: 500
47
 
48
  ### Training results
49
 
50
+ | Training Loss | Epoch | Step | Validation Loss | Micro F1 |
51
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
52
+ | 0.2722 | 4.44 | 200 | 0.0318 | 0.6422 |
53
+ | 0.0182 | 8.88 | 400 | 0.0176 | 0.8321 |
54
+ | 0.0081 | 13.33 | 600 | 0.0202 | 0.8530 |
55
+ | 0.0045 | 17.77 | 800 | 0.0168 | 0.8839 |
56
+ | 0.0029 | 22.22 | 1000 | 0.0230 | 0.8758 |
57
+ | 0.0016 | 26.66 | 1200 | 0.0234 | 0.8690 |
58
+ | 0.0013 | 31.11 | 1400 | 0.0219 | 0.8853 |
59
+ | 0.0012 | 35.55 | 1600 | 0.0248 | 0.8831 |
60
+ | 0.0008 | 39.99 | 1800 | 0.0223 | 0.8979 |
61
+ | 0.0006 | 44.44 | 2000 | 0.0257 | 0.8752 |
62
+ | 0.0004 | 48.88 | 2200 | 0.0191 | 0.9018 |
63
+ | 0.0004 | 53.33 | 2400 | 0.0244 | 0.9029 |
64
+ | 0.0004 | 57.77 | 2600 | 0.0227 | 0.8986 |
65
+ | 0.0007 | 62.22 | 2800 | 0.0242 | 0.9004 |
66
+ | 0.0004 | 66.66 | 3000 | 0.0218 | 0.8937 |
67
+ | 0.0003 | 71.11 | 3200 | 0.0214 | 0.9039 |
68
+ | 0.0003 | 75.55 | 3400 | 0.0251 | 0.8958 |
69
+ | 0.0003 | 79.99 | 3600 | 0.0210 | 0.8943 |
70
+ | 0.0004 | 84.44 | 3800 | 0.0227 | 0.8995 |
71
+ | 0.0002 | 88.88 | 4000 | 0.0226 | 0.9116 |
72
+ | 0.0002 | 93.33 | 4200 | 0.0209 | 0.9121 |
73
+ | 0.0001 | 97.77 | 4400 | 0.0225 | 0.9105 |
74
+ | 0.0003 | 102.22 | 4600 | 0.0278 | 0.9039 |
75
+ | 0.0003 | 106.66 | 4800 | 0.0223 | 0.9009 |
76
+ | 0.0002 | 111.11 | 5000 | 0.0274 | 0.9123 |
77
+ | 0.0003 | 115.55 | 5200 | 0.0236 | 0.9123 |
78
+ | 0.0001 | 119.99 | 5400 | 0.0235 | 0.9059 |
79
+ | 0.0001 | 124.44 | 5600 | 0.0220 | 0.9165 |
80
+ | 0.0002 | 128.88 | 5800 | 0.0265 | 0.9021 |
81
+ | 0.0001 | 133.33 | 6000 | 0.0225 | 0.9029 |
82
+ | 0.0002 | 137.77 | 6200 | 0.0249 | 0.9166 |
83
+ | 0.0004 | 142.22 | 6400 | 0.0192 | 0.9128 |
84
+ | 0.0002 | 146.66 | 6600 | 0.0181 | 0.9313 |
85
+ | 0.0001 | 151.11 | 6800 | 0.0191 | 0.9269 |
86
+ | 0.0001 | 155.55 | 7000 | 0.0224 | 0.9154 |
87
+ | 0.0002 | 159.99 | 7200 | 0.0192 | 0.9171 |
88
+ | 0.0002 | 164.44 | 7400 | 0.0202 | 0.9217 |
89
+ | 0.0001 | 168.88 | 7600 | 0.0185 | 0.9171 |
90
+ | 0.0001 | 173.33 | 7800 | 0.0207 | 0.9201 |
91
+ | 0.0001 | 177.77 | 8000 | 0.0226 | 0.9121 |
92
+ | 0.0001 | 182.22 | 8200 | 0.0208 | 0.9191 |
93
+ | 0.0002 | 186.66 | 8400 | 0.0248 | 0.9184 |
94
+ | 0.0002 | 191.11 | 8600 | 0.0213 | 0.9217 |
95
 
96
 
97
  ### Framework versions