Jorge Lopez Grisman commited on
Commit
869fe62
1 Parent(s): c001272

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +36 -18
README.md CHANGED
@@ -1,7 +1,7 @@
1
  ---
2
- language:
3
- - en
4
  license: apache-2.0
 
 
5
  datasets:
6
  - conll2003
7
  metrics:
@@ -10,7 +10,7 @@ metrics:
10
  - f1
11
  - accuracy
12
  model-index:
13
- - name: bert-large-uncased
14
  results:
15
  - task:
16
  name: Token Classification
@@ -18,18 +18,18 @@ model-index:
18
  dataset:
19
  name: conll2003
20
  type: conll2003
21
- args: default
22
  metrics:
23
- - name: precision
24
  type: precision
25
  value: 0.9504719600222099
26
- - name: recall
27
  type: recall
28
  value: 0.9574896520863632
29
- - name: f1
30
  type: f1
31
  value: 0.9539679001337494
32
- - name: accuracy
33
  type: accuracy
34
  value: 0.9885618059637473
35
  ---
@@ -37,14 +37,15 @@ model-index:
37
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
38
  should probably proofread and complete it, then remove this comment. -->
39
 
40
- # bert-large-uncased
41
 
42
  This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the conll2003 dataset.
43
  It achieves the following results on the evaluation set:
44
- - precision: 0.9505
45
- - recall: 0.9575
46
- - f1: 0.9540
47
- - accuracy: 0.9886
 
48
 
49
  ## Model description
50
 
@@ -63,12 +64,29 @@ More information needed
63
  ### Training hyperparameters
64
 
65
  The following hyperparameters were used during training:
66
- - num_train_epochs: 10
67
- - train_batch_size: 4
68
  - learning_rate: 2e-05
69
- - weight_decay_rate: 0.01
70
- - num_warmup_steps: 0
71
- - fp16: True
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
72
 
73
  ### Framework versions
74
 
 
1
  ---
 
 
2
  license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
  datasets:
6
  - conll2003
7
  metrics:
 
10
  - f1
11
  - accuracy
12
  model-index:
13
+ - name: bert-large-uncased-finetuned-ner
14
  results:
15
  - task:
16
  name: Token Classification
 
18
  dataset:
19
  name: conll2003
20
  type: conll2003
21
+ args: conll2003
22
  metrics:
23
+ - name: Precision
24
  type: precision
25
  value: 0.9504719600222099
26
+ - name: Recall
27
  type: recall
28
  value: 0.9574896520863632
29
+ - name: F1
30
  type: f1
31
  value: 0.9539679001337494
32
+ - name: Accuracy
33
  type: accuracy
34
  value: 0.9885618059637473
35
  ---
 
37
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
38
  should probably proofread and complete it, then remove this comment. -->
39
 
40
+ # bert-large-uncased-finetuned-ner
41
 
42
  This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the conll2003 dataset.
43
  It achieves the following results on the evaluation set:
44
+ - Loss: 0.0778
45
+ - Precision: 0.9505
46
+ - Recall: 0.9575
47
+ - F1: 0.9540
48
+ - Accuracy: 0.9886
49
 
50
  ## Model description
51
 
 
64
  ### Training hyperparameters
65
 
66
  The following hyperparameters were used during training:
 
 
67
  - learning_rate: 2e-05
68
+ - train_batch_size: 16
69
+ - eval_batch_size: 64
70
+ - seed: 42
71
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
72
+ - lr_scheduler_type: linear
73
+ - num_epochs: 10
74
+
75
+ ### Training results
76
+
77
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
78
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
79
+ | 0.1997 | 1.0 | 878 | 0.0576 | 0.9316 | 0.9257 | 0.9286 | 0.9837 |
80
+ | 0.04 | 2.0 | 1756 | 0.0490 | 0.9400 | 0.9513 | 0.9456 | 0.9870 |
81
+ | 0.0199 | 3.0 | 2634 | 0.0557 | 0.9436 | 0.9540 | 0.9488 | 0.9879 |
82
+ | 0.0112 | 4.0 | 3512 | 0.0602 | 0.9443 | 0.9569 | 0.9506 | 0.9881 |
83
+ | 0.0068 | 5.0 | 4390 | 0.0631 | 0.9451 | 0.9589 | 0.9520 | 0.9882 |
84
+ | 0.0044 | 6.0 | 5268 | 0.0638 | 0.9510 | 0.9567 | 0.9538 | 0.9885 |
85
+ | 0.003 | 7.0 | 6146 | 0.0722 | 0.9495 | 0.9560 | 0.9527 | 0.9885 |
86
+ | 0.0016 | 8.0 | 7024 | 0.0762 | 0.9491 | 0.9595 | 0.9543 | 0.9887 |
87
+ | 0.0018 | 9.0 | 7902 | 0.0769 | 0.9496 | 0.9542 | 0.9519 | 0.9883 |
88
+ | 0.0009 | 10.0 | 8780 | 0.0778 | 0.9505 | 0.9575 | 0.9540 | 0.9886 |
89
+
90
 
91
  ### Framework versions
92