Hugging Face's logo

JoshuaRubin
/
bert-base-uncased-finetuned-math_punctuation-ignore_word_parts

Token Classification
Transformers
PyTorch
TensorBoard
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
2
JoshuaRubin commited on
Commit
13b6dbb
•
1 Parent(s): 1aff953

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -24
README.md CHANGED
@@ -17,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.1788
21
- - Precision: 0.7968
22
- - Recall: 0.7772
23
- - F Score: 0.7856
24
- - Auc: 0.9369
25
 
26
  ## Model description
27
 
@@ -52,25 +52,24 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F Score | Auc |
54
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:-------:|:------:|
55
- | 0.066 | 0.61 | 500 | 0.0912 | 0.7938 | 0.7091 | 0.7337 | 0.9273 |
56
- | 0.0586 | 1.23 | 1000 | 0.0954 | 0.7823 | 0.7593 | 0.7684 | 0.9350 |
57
- | 0.0548 | 1.84 | 1500 | 0.0866 | 0.7860 | 0.7616 | 0.7722 | 0.9431 |
58
- | 0.0375 | 2.45 | 2000 | 0.0953 | 0.7834 | 0.7696 | 0.7757 | 0.9382 |
59
- | 0.0348 | 3.07 | 2500 | 0.1051 | 0.7803 | 0.7767 | 0.7784 | 0.9403 |
60
- | 0.025 | 3.68 | 3000 | 0.1050 | 0.7963 | 0.7677 | 0.7789 | 0.9397 |
61
- | 0.0166 | 4.29 | 3500 | 0.1182 | 0.7846 | 0.7731 | 0.7777 | 0.9384 |
62
- | 0.0169 | 4.91 | 4000 | 0.1230 | 0.7893 | 0.7607 | 0.7730 | 0.9377 |
63
- | 0.0099 | 5.52 | 4500 | 0.1422 | 0.7872 | 0.7793 | 0.7812 | 0.9366 |
64
- | 0.0096 | 6.13 | 5000 | 0.1451 | 0.7954 | 0.7747 | 0.7833 | 0.9369 |
65
- | 0.0074 | 6.75 | 5500 | 0.1458 | 0.7809 | 0.7747 | 0.7765 | 0.9394 |
66
- | 0.0053 | 7.36 | 6000 | 0.1610 | 0.7929 | 0.7643 | 0.7760 | 0.9342 |
67
- | 0.0051 | 7.98 | 6500 | 0.1540 | 0.7893 | 0.7735 | 0.7810 | 0.9377 |
68
- | 0.0038 | 8.59 | 7000 | 0.1633 | 0.7800 | 0.7759 | 0.7770 | 0.9358 |
69
- | 0.0031 | 9.2 | 7500 | 0.1665 | 0.7866 | 0.7736 | 0.7782 | 0.9380 |
70
- | 0.0026 | 9.82 | 8000 | 0.1750 | 0.7924 | 0.7798 | 0.7854 | 0.9359 |
71
- | 0.0018 | 10.43 | 8500 | 0.1732 | 0.7967 | 0.7754 | 0.7842 | 0.9380 |
72
- | 0.0016 | 11.04 | 9000 | 0.1780 | 0.7969 | 0.7757 | 0.7835 | 0.9373 |
73
- | 0.0015 | 11.66 | 9500 | 0.1788 | 0.7968 | 0.7772 | 0.7856 | 0.9369 |
74
 
75
 
76
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.1981
21
+ - Precision: 0.7843
22
+ - Recall: 0.7485
23
+ - F Score: 0.7648
24
+ - Auc: 0.9248
25
 
26
  ## Model description
27
 
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F Score | Auc |
54
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:-------:|:------:|
55
+ | 0.1064 | 0.64 | 500 | 0.1082 | 0.7558 | 0.6580 | 0.6964 | 0.9086 |
56
+ | 0.0781 | 1.27 | 1000 | 0.1025 | 0.7594 | 0.7226 | 0.7365 | 0.9261 |
57
+ | 0.0757 | 1.91 | 1500 | 0.1001 | 0.7945 | 0.6899 | 0.7302 | 0.9272 |
58
+ | 0.0538 | 2.54 | 2000 | 0.1061 | 0.7689 | 0.7348 | 0.7480 | 0.9298 |
59
+ | 0.0425 | 3.18 | 2500 | 0.1123 | 0.7806 | 0.7361 | 0.7560 | 0.9300 |
60
+ | 0.0377 | 3.81 | 3000 | 0.1159 | 0.7841 | 0.7437 | 0.7610 | 0.9292 |
61
+ | 0.0235 | 4.45 | 3500 | 0.1259 | 0.7786 | 0.7368 | 0.7561 | 0.9276 |
62
+ | 0.0227 | 5.08 | 4000 | 0.1436 | 0.7699 | 0.7448 | 0.7555 | 0.9277 |
63
+ | 0.0159 | 5.72 | 4500 | 0.1466 | 0.7715 | 0.7333 | 0.7514 | 0.9252 |
64
+ | 0.0106 | 6.35 | 5000 | 0.1574 | 0.7710 | 0.7456 | 0.7566 | 0.9276 |
65
+ | 0.0111 | 6.99 | 5500 | 0.1560 | 0.7694 | 0.7500 | 0.7595 | 0.9286 |
66
+ | 0.0074 | 7.62 | 6000 | 0.1645 | 0.7789 | 0.7511 | 0.7639 | 0.9305 |
67
+ | 0.0056 | 8.26 | 6500 | 0.1745 | 0.7887 | 0.7453 | 0.7648 | 0.9265 |
68
+ | 0.005 | 8.89 | 7000 | 0.1760 | 0.7779 | 0.7497 | 0.7629 | 0.9281 |
69
+ | 0.0038 | 9.53 | 7500 | 0.1873 | 0.7826 | 0.7505 | 0.7634 | 0.9273 |
70
+ | 0.0031 | 10.17 | 8000 | 0.1896 | 0.7855 | 0.7477 | 0.7644 | 0.9258 |
71
+ | 0.0026 | 10.8 | 8500 | 0.1929 | 0.7849 | 0.7485 | 0.7650 | 0.9263 |
72
+ | 0.0017 | 11.44 | 9000 | 0.1981 | 0.7843 | 0.7485 | 0.7648 | 0.9248 |
 
73
 
74
 
75
  ### Framework versions