Hugging Face's logo

CIRCL
/
vulnerability-severity-classification-chinese-macbert-base-test

Text Classification
Transformers
Safetensors
Chinese
bert
vulnerability
severity
cybersecurity
cnvd
text-embeddings-inference
Model card Files
xet
 Community
cedricbonhomme commited on
Commit
4da6a28
·
verified ·
1 Parent(s): 1f603f1

End of training

Browse files
Files changed (3) hide show
  1. README.md +21 -21
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -18,18 +18,18 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [hfl/chinese-macbert-base](https://huggingface.co/hfl/chinese-macbert-base) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.6835
22
- - Accuracy: 0.7457
23
- - F1 Macro: 0.6781
24
- - Low Precision: 0.5036
25
- - Low Recall: 0.4902
26
- - Low F1: 0.4968
27
- - Medium Precision: 0.8005
28
- - Medium Recall: 0.7481
29
- - Medium F1: 0.7734
30
- - High Precision: 0.7266
31
- - High Recall: 0.8056
32
- - High F1: 0.7640
33
 
34
  ## Model description
35
 
@@ -49,8 +49,8 @@ More information needed
49
 
50
  The following hyperparameters were used during training:
51
  - learning_rate: 3e-05
52
- - train_batch_size: 64
53
- - eval_batch_size: 64
54
  - seed: 42
55
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
56
  - lr_scheduler_type: linear
@@ -58,13 +58,13 @@ The following hyperparameters were used during training:
58
 
59
  ### Training results
60
 
61
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Low Precision | Low Recall | Low F1 | Medium Precision | Medium Recall | Medium F1 | High Precision | High Recall | High F1 |
62
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-------------:|:----------:|:------:|:----------------:|:-------------:|:---------:|:--------------:|:-----------:|:-------:|
63
- | 0.6748 | 1.0 | 1590 | 0.7081 | 0.7197 | 0.6421 | 0.4647 | 0.4108 | 0.4361 | 0.7747 | 0.7230 | 0.7480 | 0.6987 | 0.7915 | 0.7422 |
64
- | 0.7115 | 2.0 | 3180 | 0.6628 | 0.7406 | 0.6677 | 0.4907 | 0.4579 | 0.4737 | 0.7895 | 0.7507 | 0.7696 | 0.7273 | 0.7952 | 0.7598 |
65
- | 0.5559 | 3.0 | 4770 | 0.6606 | 0.7293 | 0.6682 | 0.4384 | 0.5573 | 0.4908 | 0.8111 | 0.6990 | 0.7509 | 0.7138 | 0.8194 | 0.7629 |
66
- | 0.5732 | 4.0 | 6360 | 0.6653 | 0.7394 | 0.6756 | 0.4559 | 0.5416 | 0.4951 | 0.8066 | 0.7271 | 0.7648 | 0.7298 | 0.8080 | 0.7669 |
67
- | 0.5008 | 5.0 | 7950 | 0.6835 | 0.7457 | 0.6781 | 0.5036 | 0.4902 | 0.4968 | 0.8005 | 0.7481 | 0.7734 | 0.7266 | 0.8056 | 0.7640 |
68
 
69
 
70
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [hfl/chinese-macbert-base](https://huggingface.co/hfl/chinese-macbert-base) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 1.9271
22
+ - Accuracy: 0.7647
23
+ - F1 Macro: 0.6842
24
+ - Low Precision: 0.6080
25
+ - Low Recall: 0.4038
26
+ - Low F1: 0.4853
27
+ - Medium Precision: 0.7836
28
+ - Medium Recall: 0.8137
29
+ - Medium F1: 0.7984
30
+ - High Precision: 0.7600
31
+ - High Recall: 0.7782
32
+ - High F1: 0.7690
33
 
34
  ## Model description
35
 
 
49
 
50
  The following hyperparameters were used during training:
51
  - learning_rate: 3e-05
52
+ - train_batch_size: 48
53
+ - eval_batch_size: 48
54
  - seed: 42
55
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
56
  - lr_scheduler_type: linear
 
58
 
59
  ### Training results
60
 
61
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Low Precision | Low Recall | Low F1 | Medium Precision | Medium Recall | Medium F1 | High Precision | High Recall | High F1 |
62
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:-------------:|:----------:|:------:|:----------------:|:-------------:|:---------:|:--------------:|:-----------:|:-------:|
63
+ | 1.7246 | 1.0 | 2120 | 1.8115 | 0.7468 | 0.6062 | 0.6827 | 0.1736 | 0.2768 | 0.7552 | 0.8203 | 0.7864 | 0.7370 | 0.7749 | 0.7555 |
64
+ | 1.7560 | 2.0 | 4240 | 1.6959 | 0.7649 | 0.6670 | 0.6188 | 0.3362 | 0.4357 | 0.7580 | 0.8613 | 0.8064 | 0.8006 | 0.7214 | 0.7590 |
65
+ | 1.2407 | 3.0 | 6360 | 1.7663 | 0.7664 | 0.6825 | 0.6071 | 0.3894 | 0.4745 | 0.7895 | 0.8096 | 0.7994 | 0.7554 | 0.7930 | 0.7737 |
66
+ | 1.2336 | 4.0 | 8480 | 1.8252 | 0.7678 | 0.6870 | 0.5968 | 0.4099 | 0.4860 | 0.7866 | 0.8167 | 0.8014 | 0.7665 | 0.7806 | 0.7735 |
67
+ | 1.0107 | 5.0 | 10600 | 1.9271 | 0.7647 | 0.6842 | 0.6080 | 0.4038 | 0.4853 | 0.7836 | 0.8137 | 0.7984 | 0.7600 | 0.7782 | 0.7690 |
68
 
69
 
70
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6d51c981eb35042238c02508e07e9b342c75049f4ca2f7469e720c758e18c51e
3
  size 409103292
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e41fa0a0cf912511fdf4c74c65f599329793b3130d8530d5418f9760924f8854
3
  size 409103292
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:841221ed0c084667e4dde4275f8b6c281a157b9d51a0da45e3fe9ffb046c2ac2
3
  size 5265
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1772bd105fbd572b97a68871eadbcff379020c8bf2bc9c9ae2f05d39342fdec5
3
  size 5265