Hugging Face's logo

HYM
/
Cbert_base_ws-finetuned-ner

Token Classification
Transformers
PyTorch
bert
Generated from Trainer
Model card Files
xet
Community
1
HYM commited on
Commit
fca6a0f
·
1 Parent(s): 65d2d70

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -26
README.md CHANGED
@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [ckiplab/bert-base-chinese-ws](https://huggingface.co/ckiplab/bert-base-chinese-ws) on an unknown dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.1284
23
- - Precision: 0.9594
24
- - Recall: 0.9636
25
- - F1: 0.9615
26
- - Accuracy: 0.9825
27
 
28
  ## Model description
29
 
@@ -48,32 +48,27 @@ The following hyperparameters were used during training:
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
- - num_epochs: 20
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
- | 0.0435 | 1.0 | 1565 | 0.0503 | 0.9566 | 0.9606 | 0.9586 | 0.9813 |
58
- | 0.0335 | 2.0 | 3130 | 0.0500 | 0.9613 | 0.9636 | 0.9624 | 0.9830 |
59
- | 0.0256 | 3.0 | 4695 | 0.0544 | 0.9576 | 0.9639 | 0.9607 | 0.9822 |
60
- | 0.0206 | 4.0 | 6260 | 0.0565 | 0.9624 | 0.9622 | 0.9623 | 0.9830 |
61
- | 0.0167 | 5.0 | 7825 | 0.0689 | 0.9610 | 0.9627 | 0.9619 | 0.9827 |
62
- | 0.0137 | 6.0 | 9390 | 0.0733 | 0.9597 | 0.9632 | 0.9614 | 0.9825 |
63
- | 0.0112 | 7.0 | 10955 | 0.0774 | 0.9594 | 0.9628 | 0.9611 | 0.9823 |
64
- | 0.0092 | 8.0 | 12520 | 0.0833 | 0.9587 | 0.9622 | 0.9604 | 0.9821 |
65
- | 0.0083 | 9.0 | 14085 | 0.0942 | 0.9577 | 0.9627 | 0.9602 | 0.9820 |
66
- | 0.007 | 10.0 | 15650 | 0.0925 | 0.9597 | 0.9625 | 0.9611 | 0.9824 |
67
- | 0.0066 | 11.0 | 17215 | 0.1021 | 0.9587 | 0.9618 | 0.9602 | 0.9820 |
68
- | 0.0056 | 12.0 | 18780 | 0.1057 | 0.9585 | 0.9618 | 0.9602 | 0.9820 |
69
- | 0.005 | 13.0 | 20345 | 0.1076 | 0.9590 | 0.9629 | 0.9609 | 0.9823 |
70
- | 0.0042 | 14.0 | 21910 | 0.1089 | 0.9583 | 0.9628 | 0.9605 | 0.9820 |
71
- | 0.0041 | 15.0 | 23475 | 0.1108 | 0.9596 | 0.9624 | 0.9610 | 0.9823 |
72
- | 0.0042 | 16.0 | 25040 | 0.1172 | 0.9592 | 0.9631 | 0.9612 | 0.9823 |
73
- | 0.0036 | 17.0 | 26605 | 0.1153 | 0.9598 | 0.9628 | 0.9613 | 0.9825 |
74
- | 0.003 | 18.0 | 28170 | 0.1225 | 0.9601 | 0.9634 | 0.9618 | 0.9827 |
75
- | 0.0029 | 19.0 | 29735 | 0.1247 | 0.9603 | 0.9630 | 0.9617 | 0.9826 |
76
- | 0.0029 | 20.0 | 31300 | 0.1284 | 0.9594 | 0.9636 | 0.9615 | 0.9825 |
77
 
78
 
79
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [ckiplab/bert-base-chinese-ws](https://huggingface.co/ckiplab/bert-base-chinese-ws) on an unknown dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.0582
23
+ - Precision: 0.9602
24
+ - Recall: 0.9633
25
+ - F1: 0.9617
26
+ - Accuracy: 0.9827
27
 
28
  ## Model description
29
 
 
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
+ - num_epochs: 10
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
+ | 0.0482 | 0.64 | 1000 | 0.0509 | 0.9601 | 0.9582 | 0.9592 | 0.9817 |
58
+ | 0.0364 | 1.28 | 2000 | 0.0521 | 0.9590 | 0.9615 | 0.9602 | 0.9820 |
59
+ | 0.0341 | 1.92 | 3000 | 0.0548 | 0.9546 | 0.9625 | 0.9585 | 0.9812 |
60
+ | 0.0264 | 2.56 | 4000 | 0.0550 | 0.9593 | 0.9623 | 0.9608 | 0.9822 |
61
+ | 0.0227 | 3.19 | 5000 | 0.0582 | 0.9602 | 0.9633 | 0.9617 | 0.9827 |
62
+ | 0.021 | 3.83 | 6000 | 0.0595 | 0.9581 | 0.9624 | 0.9603 | 0.9820 |
63
+ | 0.0162 | 4.47 | 7000 | 0.0686 | 0.9574 | 0.9626 | 0.9600 | 0.9819 |
64
+ | 0.0159 | 5.11 | 8000 | 0.0719 | 0.9596 | 0.9614 | 0.9605 | 0.9822 |
65
+ | 0.0144 | 5.75 | 9000 | 0.0732 | 0.9590 | 0.9620 | 0.9605 | 0.9822 |
66
+ | 0.0109 | 6.39 | 10000 | 0.0782 | 0.9599 | 0.9626 | 0.9612 | 0.9824 |
67
+ | 0.0122 | 7.03 | 11000 | 0.0803 | 0.9605 | 0.9620 | 0.9612 | 0.9825 |
68
+ | 0.0097 | 7.67 | 12000 | 0.0860 | 0.9591 | 0.9620 | 0.9605 | 0.9822 |
69
+ | 0.0087 | 8.31 | 13000 | 0.0877 | 0.9591 | 0.9616 | 0.9603 | 0.9821 |
70
+ | 0.0087 | 8.95 | 14000 | 0.0902 | 0.9585 | 0.9630 | 0.9607 | 0.9823 |
71
+ | 0.0078 | 9.58 | 15000 | 0.0929 | 0.9589 | 0.9621 | 0.9605 | 0.9821 |
 
 
 
 
 
72
 
73
 
74
  ### Framework versions