Hugging Face's logo

luohuashijieyoufengjun
/
ner_based_bert-base-chinese

Token Classification
Transformers
Safetensors
bert
Generated from Trainer
Model card Files
xet
 Community
luohuashijieyoufengjun commited on
Commit
8d3835c
·
verified ·
1 Parent(s): ee56ad9

End of training

Browse files
Files changed (4) hide show
  1. README.md +24 -24
  2. config.json +1 -2
  3. model.safetensors +1 -1
  4. training_args.bin +2 -2
README.md CHANGED
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [google-bert/bert-base-chinese](https://huggingface.co/google-bert/bert-base-chinese) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.0179
24
- - Precision: 0.9608
25
- - Recall: 0.9684
26
- - F1: 0.9646
27
- - Accuracy: 0.9972
28
 
29
  ## Model description
30
 
@@ -56,26 +56,26 @@ The following hyperparameters were used during training:
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
58
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
59
- | 0.1098 | 1.0 | 650 | 0.0206 | 0.9073 | 0.9340 | 0.9205 | 0.9941 |
60
- | 0.0199 | 2.0 | 1300 | 0.0164 | 0.9308 | 0.9510 | 0.9408 | 0.9955 |
61
- | 0.0148 | 3.0 | 1950 | 0.0144 | 0.9425 | 0.9573 | 0.9498 | 0.9963 |
62
- | 0.0086 | 4.0 | 2600 | 0.0142 | 0.9476 | 0.9621 | 0.9548 | 0.9965 |
63
- | 0.007 | 5.0 | 3250 | 0.0144 | 0.9526 | 0.9645 | 0.9585 | 0.9967 |
64
- | 0.0056 | 6.0 | 3900 | 0.0144 | 0.9514 | 0.9686 | 0.9599 | 0.9969 |
65
- | 0.0037 | 7.0 | 4550 | 0.0158 | 0.9539 | 0.9677 | 0.9608 | 0.9968 |
66
- | 0.003 | 8.0 | 5200 | 0.0156 | 0.9560 | 0.9677 | 0.9618 | 0.9970 |
67
- | 0.0026 | 9.0 | 5850 | 0.0164 | 0.9521 | 0.9699 | 0.9609 | 0.9969 |
68
- | 0.0019 | 10.0 | 6500 | 0.0157 | 0.9595 | 0.9678 | 0.9636 | 0.9971 |
69
- | 0.0015 | 11.0 | 7150 | 0.0162 | 0.9601 | 0.9664 | 0.9633 | 0.9971 |
70
- | 0.0014 | 12.0 | 7800 | 0.0176 | 0.9590 | 0.9673 | 0.9631 | 0.9971 |
71
- | 0.0013 | 13.0 | 8450 | 0.0179 | 0.9588 | 0.9681 | 0.9634 | 0.9971 |
72
- | 0.001 | 14.0 | 9100 | 0.0180 | 0.9605 | 0.9693 | 0.9649 | 0.9972 |
73
- | 0.0009 | 15.0 | 9750 | 0.0179 | 0.9608 | 0.9684 | 0.9646 | 0.9972 |
74
 
75
 
76
  ### Framework versions
77
 
78
- - Transformers 4.48.3
79
- - Pytorch 2.6.0+cu126
80
- - Datasets 3.2.0
81
- - Tokenizers 0.21.0
 
20
 
21
  This model is a fine-tuned version of [google-bert/bert-base-chinese](https://huggingface.co/google-bert/bert-base-chinese) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.0171
24
+ - Precision: 0.9610
25
+ - Recall: 0.9716
26
+ - F1: 0.9663
27
+ - Accuracy: 0.9973
28
 
29
  ## Model description
30
 
 
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
58
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
59
+ | 0.0941 | 1.0 | 650 | 0.0194 | 0.9150 | 0.9327 | 0.9237 | 0.9943 |
60
+ | 0.0193 | 2.0 | 1300 | 0.0160 | 0.9282 | 0.9546 | 0.9412 | 0.9954 |
61
+ | 0.0149 | 3.0 | 1950 | 0.0142 | 0.9477 | 0.9577 | 0.9527 | 0.9964 |
62
+ | 0.0088 | 4.0 | 2600 | 0.0128 | 0.9551 | 0.9604 | 0.9577 | 0.9967 |
63
+ | 0.0069 | 5.0 | 3250 | 0.0135 | 0.9567 | 0.9635 | 0.9601 | 0.9968 |
64
+ | 0.0056 | 6.0 | 3900 | 0.0134 | 0.9552 | 0.9669 | 0.9610 | 0.9970 |
65
+ | 0.0037 | 7.0 | 4550 | 0.0137 | 0.9592 | 0.9688 | 0.9640 | 0.9971 |
66
+ | 0.0031 | 8.0 | 5200 | 0.0144 | 0.9592 | 0.9673 | 0.9632 | 0.9971 |
67
+ | 0.0026 | 9.0 | 5850 | 0.0157 | 0.9536 | 0.9711 | 0.9623 | 0.9970 |
68
+ | 0.0019 | 10.0 | 6500 | 0.0159 | 0.9586 | 0.9706 | 0.9646 | 0.9971 |
69
+ | 0.0016 | 11.0 | 7150 | 0.0163 | 0.9592 | 0.9711 | 0.9651 | 0.9972 |
70
+ | 0.0015 | 12.0 | 7800 | 0.0164 | 0.9621 | 0.9702 | 0.9661 | 0.9972 |
71
+ | 0.0013 | 13.0 | 8450 | 0.0166 | 0.9625 | 0.9714 | 0.9669 | 0.9973 |
72
+ | 0.001 | 14.0 | 9100 | 0.0171 | 0.9624 | 0.9711 | 0.9667 | 0.9973 |
73
+ | 0.0009 | 15.0 | 9750 | 0.0171 | 0.9610 | 0.9716 | 0.9663 | 0.9973 |
74
 
75
 
76
  ### Framework versions
77
 
78
+ - Transformers 4.51.3
79
+ - Pytorch 2.7.0+cu126
80
+ - Datasets 3.6.0
81
+ - Tokenizers 0.21.1
config.json CHANGED
@@ -1,5 +1,4 @@
1
  {
2
- "_name_or_path": "google-bert/bert-base-chinese",
3
  "architectures": [
4
  "BertForTokenClassification"
5
  ],
@@ -72,7 +71,7 @@
72
  "pooler_type": "first_token_transform",
73
  "position_embedding_type": "absolute",
74
  "torch_dtype": "float32",
75
- "transformers_version": "4.48.3",
76
  "type_vocab_size": 2,
77
  "use_cache": true,
78
  "vocab_size": 21128
 
1
  {
 
2
  "architectures": [
3
  "BertForTokenClassification"
4
  ],
 
71
  "pooler_type": "first_token_transform",
72
  "position_embedding_type": "absolute",
73
  "torch_dtype": "float32",
74
+ "transformers_version": "4.51.3",
75
  "type_vocab_size": 2,
76
  "use_cache": true,
77
  "vocab_size": 21128
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2d83c8b6be6a5be5e0d643debbe46c6680be922fae25c675c202033006c6f6ba
3
  size 406799208
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0cb6fdc369784e1c5c2f47b17081f2e0f34da2dbaf0c8f592763ef8535bd608
3
  size 406799208
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e9abbe4c4643c65c84952e7955ec58231e872affd82e3b0e8652350e783ca2f6
3
- size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ffb872709eec0a95eed970df3306e01b69a36c60a206ad08474daa9f15187a7
3
+ size 5713