Hugging Face's logo

Tommert25
/
multibert_dataaugmentation

Token Classification
Transformers
PyTorch
bert
Generated from Trainer
Inference Endpoints
Model card Files Community
Tommert25 commited on
Commit
acb3321
·
1 Parent(s): b1b524c

End of training

Browse files
Files changed (2) hide show
  1. README.md +19 -19
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [bert-base-multilingual-uncased](https://huggingface.co/bert-base-multilingual-uncased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.5981
22
- - Precisions: 0.8870
23
- - Recall: 0.8815
24
- - F-measure: 0.8825
25
- - Accuracy: 0.9265
26
 
27
  ## Model description
28
 
@@ -53,20 +53,20 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Precisions | Recall | F-measure | Accuracy |
55
  |:-------------:|:-----:|:----:|:---------------:|:----------:|:------:|:---------:|:--------:|
56
- | 0.5651 | 1.0 | 284 | 0.4671 | 0.8381 | 0.7291 | 0.7488 | 0.8699 |
57
- | 0.2863 | 2.0 | 568 | 0.4120 | 0.8143 | 0.8209 | 0.8151 | 0.8946 |
58
- | 0.1717 | 3.0 | 852 | 0.3688 | 0.8778 | 0.8430 | 0.8579 | 0.9065 |
59
- | 0.1085 | 4.0 | 1136 | 0.4175 | 0.8609 | 0.8656 | 0.8624 | 0.9097 |
60
- | 0.0677 | 5.0 | 1420 | 0.4885 | 0.8728 | 0.8483 | 0.8588 | 0.9142 |
61
- | 0.0526 | 6.0 | 1704 | 0.5384 | 0.8672 | 0.8786 | 0.8672 | 0.9152 |
62
- | 0.0372 | 7.0 | 1988 | 0.4872 | 0.8766 | 0.8770 | 0.8759 | 0.9161 |
63
- | 0.0239 | 8.0 | 2272 | 0.5377 | 0.9011 | 0.8665 | 0.8799 | 0.9227 |
64
- | 0.0181 | 9.0 | 2556 | 0.5604 | 0.8846 | 0.8796 | 0.8813 | 0.9236 |
65
- | 0.0117 | 10.0 | 2840 | 0.5815 | 0.8695 | 0.8786 | 0.8700 | 0.9229 |
66
- | 0.008 | 11.0 | 3124 | 0.6035 | 0.8842 | 0.8701 | 0.8744 | 0.9206 |
67
- | 0.0047 | 12.0 | 3408 | 0.5789 | 0.8644 | 0.8815 | 0.8662 | 0.9232 |
68
- | 0.0037 | 13.0 | 3692 | 0.5891 | 0.8763 | 0.8782 | 0.8740 | 0.9259 |
69
- | 0.0026 | 14.0 | 3976 | 0.5981 | 0.8870 | 0.8815 | 0.8825 | 0.9265 |
70
 
71
 
72
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [bert-base-multilingual-uncased](https://huggingface.co/bert-base-multilingual-uncased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.7216
22
+ - Precisions: 0.8745
23
+ - Recall: 0.8029
24
+ - F-measure: 0.8291
25
+ - Accuracy: 0.9015
26
 
27
  ## Model description
28
 
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Precisions | Recall | F-measure | Accuracy |
55
  |:-------------:|:-----:|:----:|:---------------:|:----------:|:------:|:---------:|:--------:|
56
+ | 0.5776 | 1.0 | 284 | 0.5209 | 0.7928 | 0.6866 | 0.7216 | 0.8385 |
57
+ | 0.2774 | 2.0 | 568 | 0.4278 | 0.8274 | 0.7457 | 0.7626 | 0.8802 |
58
+ | 0.1471 | 3.0 | 852 | 0.5814 | 0.8455 | 0.7114 | 0.7399 | 0.8627 |
59
+ | 0.0933 | 4.0 | 1136 | 0.5386 | 0.8006 | 0.7300 | 0.7556 | 0.8835 |
60
+ | 0.0631 | 5.0 | 1420 | 0.6197 | 0.8122 | 0.7445 | 0.7682 | 0.8846 |
61
+ | 0.0437 | 6.0 | 1704 | 0.6973 | 0.8675 | 0.7423 | 0.7802 | 0.8815 |
62
+ | 0.0274 | 7.0 | 1988 | 0.6409 | 0.7960 | 0.7586 | 0.7729 | 0.8895 |
63
+ | 0.0202 | 8.0 | 2272 | 0.6477 | 0.8550 | 0.7712 | 0.8003 | 0.8922 |
64
+ | 0.0136 | 9.0 | 2556 | 0.7231 | 0.8722 | 0.7765 | 0.8088 | 0.8919 |
65
+ | 0.0094 | 10.0 | 2840 | 0.6787 | 0.8585 | 0.7916 | 0.8106 | 0.8955 |
66
+ | 0.0064 | 11.0 | 3124 | 0.7215 | 0.8750 | 0.7819 | 0.8107 | 0.8974 |
67
+ | 0.0052 | 12.0 | 3408 | 0.7295 | 0.8811 | 0.7840 | 0.8154 | 0.8985 |
68
+ | 0.0038 | 13.0 | 3692 | 0.7216 | 0.8745 | 0.8029 | 0.8291 | 0.9015 |
69
+ | 0.003 | 14.0 | 3976 | 0.7180 | 0.8745 | 0.8023 | 0.8289 | 0.9001 |
70
 
71
 
72
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8f3a277d980d49e6cdac7079d8cc3085d449bf9615a562ae62a8840dfa09a01e
3
  size 667152553
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:709eac0c53e1483e7bceb772bb34c91107651d3e3d2d34506033e1a96731ff27
3
  size 667152553