Hugging Face's logo

Salvatore
/
bert-finetuned-mutation-recognition-2

Token Classification
Transformers
PyTorch
TensorBoard
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
1
Salvatore commited on
Commit
5644651
•
1 Parent(s): f813834

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -20
README.md CHANGED
@@ -19,14 +19,14 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.0804
23
- - Dnamutation F1: 0.6706
24
- - Proteinmutation F1: 0.8276
25
- - Snp F1: 0.0455
26
- - Precision: 0.7373
27
- - Recall: 0.6948
28
- - F1: 0.7154
29
- - Accuracy: 0.9874
30
 
31
  ## Model description
32
 
@@ -55,18 +55,18 @@ The following hyperparameters were used during training:
55
 
56
  ### Training results
57
 
58
- | Training Loss | Epoch | Step | Validation Loss | Dnamutation F1 | Proteinmutation F1 | Snp F1 | Precision | Recall | F1 | Accuracy |
59
- |:-------------:|:-----:|:----:|:---------------:|:--------------:|:------------------:|:------:|:---------:|:------:|:------:|:--------:|
60
- | No log | 1.0 | 403 | 0.0433 | 0.4712 | 0.75 | 0.0 | 0.5581 | 0.6180 | 0.5865 | 0.9858 |
61
- | 0.085 | 2.0 | 806 | 0.0461 | 0.5506 | 0.7673 | 0.0 | 0.6454 | 0.6219 | 0.6334 | 0.9856 |
62
- | 0.0309 | 3.0 | 1209 | 0.0519 | 0.5022 | 0.7830 | 0.0 | 0.6781 | 0.5701 | 0.6194 | 0.9850 |
63
- | 0.0188 | 4.0 | 1612 | 0.0452 | 0.6008 | 0.8180 | 0.1892 | 0.6757 | 0.6718 | 0.6737 | 0.9863 |
64
- | 0.0128 | 5.0 | 2015 | 0.0494 | 0.6753 | 0.7974 | 0.0 | 0.6975 | 0.7083 | 0.7029 | 0.9870 |
65
- | 0.0128 | 6.0 | 2418 | 0.0519 | 0.6654 | 0.8421 | 0.0645 | 0.7017 | 0.7179 | 0.7097 | 0.9881 |
66
- | 0.007 | 7.0 | 2821 | 0.0745 | 0.5954 | 0.8055 | 0.2353 | 0.7072 | 0.6257 | 0.6640 | 0.9859 |
67
- | 0.0037 | 8.0 | 3224 | 0.0717 | 0.6882 | 0.8486 | 0.0 | 0.7280 | 0.7294 | 0.7287 | 0.9874 |
68
- | 0.0019 | 9.0 | 3627 | 0.0839 | 0.6373 | 0.8089 | 0.0 | 0.7209 | 0.6545 | 0.6861 | 0.9867 |
69
- | 0.0012 | 10.0 | 4030 | 0.0804 | 0.6706 | 0.8276 | 0.0455 | 0.7373 | 0.6948 | 0.7154 | 0.9874 |
70
 
71
 
72
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.0818
23
+ - Dnamutation F1: 0.6371
24
+ - Snp F1: 0.0952
25
+ - Proteinmutation F1: 0.8412
26
+ - Precision: 0.7646
27
+ - Recall: 0.6596
28
+ - F1: 0.7082
29
+ - Accuracy: 0.9877
30
 
31
  ## Model description
32
 
 
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Dnamutation F1 | Snp F1 | Proteinmutation F1 | Precision | Recall | F1 | Accuracy |
59
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:------:|:------------------:|:---------:|:------:|:------:|:--------:|
60
+ | No log | 1.0 | 403 | 0.0383 | 0.5871 | 0.0 | 0.7573 | 0.6195 | 0.6770 | 0.6470 | 0.9872 |
61
+ | 0.0863 | 2.0 | 806 | 0.0349 | 0.6202 | 0.0 | 0.8646 | 0.6815 | 0.7408 | 0.7099 | 0.9889 |
62
+ | 0.0295 | 3.0 | 1209 | 0.0415 | 0.5670 | 0.0 | 0.7689 | 0.6887 | 0.6035 | 0.6433 | 0.9866 |
63
+ | 0.019 | 4.0 | 1612 | 0.0430 | 0.5909 | 0.4742 | 0.7840 | 0.6667 | 0.6615 | 0.6641 | 0.9881 |
64
+ | 0.0127 | 5.0 | 2015 | 0.0507 | 0.6345 | 0.0 | 0.8455 | 0.7290 | 0.6867 | 0.7072 | 0.9885 |
65
+ | 0.0127 | 6.0 | 2418 | 0.0678 | 0.5946 | 0.05 | 0.8087 | 0.7471 | 0.6170 | 0.6758 | 0.9868 |
66
+ | 0.0067 | 7.0 | 2821 | 0.0544 | 0.6693 | 0.2727 | 0.8475 | 0.7208 | 0.7292 | 0.725 | 0.9884 |
67
+ | 0.0042 | 8.0 | 3224 | 0.0642 | 0.6694 | 0.2000 | 0.8401 | 0.7390 | 0.7118 | 0.7251 | 0.9885 |
68
+ | 0.0019 | 9.0 | 3627 | 0.0847 | 0.6271 | 0.0976 | 0.8416 | 0.7671 | 0.6499 | 0.7037 | 0.9877 |
69
+ | 0.0014 | 10.0 | 4030 | 0.0818 | 0.6371 | 0.0952 | 0.8412 | 0.7646 | 0.6596 | 0.7082 | 0.9877 |
70
 
71
 
72
  ### Framework versions