Hugging Face's logo

mor40
/
BulBERT-ner-wikiann

Token Classification
Transformers
PyTorch
bert
Generated from Trainer
Model card Files Community
mor40 commited on
Commit
678b326
·
1 Parent(s): db801fc

Training complete

Browse files
Files changed (1) hide show
  1. README.md +13 -17
README.md CHANGED
@@ -1,4 +1,5 @@
1
  ---
 
2
  tags:
3
  - generated_from_trainer
4
  metrics:
@@ -16,13 +17,13 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # BulBERT-ner-wikiann
18
 
19
- This model was trained from scratch on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.5594
22
- - Precision: 0.8396
23
- - Recall: 0.8739
24
- - F1: 0.8564
25
- - Accuracy: 0.9465
26
 
27
  ## Model description
28
 
@@ -47,22 +48,17 @@ The following hyperparameters were used during training:
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
- - num_epochs: 10
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
55
  |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
56
- | 0.0129 | 1.0 | 2030 | 0.4952 | 0.8126 | 0.8633 | 0.8372 | 0.9432 |
57
- | 0.0102 | 2.0 | 4060 | 0.5372 | 0.8136 | 0.8656 | 0.8388 | 0.9409 |
58
- | 0.0106 | 3.0 | 6090 | 0.5325 | 0.8170 | 0.8636 | 0.8396 | 0.9415 |
59
- | 0.0065 | 4.0 | 8120 | 0.5319 | 0.8318 | 0.8635 | 0.8473 | 0.9449 |
60
- | 0.0096 | 5.0 | 10150 | 0.5248 | 0.8271 | 0.8659 | 0.8460 | 0.9455 |
61
- | 0.004 | 6.0 | 12180 | 0.5376 | 0.8394 | 0.8703 | 0.8546 | 0.9463 |
62
- | 0.001 | 7.0 | 14210 | 0.5530 | 0.8319 | 0.8677 | 0.8494 | 0.9451 |
63
- | 0.0009 | 8.0 | 16240 | 0.5511 | 0.8362 | 0.8748 | 0.8551 | 0.9458 |
64
- | 0.0005 | 9.0 | 18270 | 0.5637 | 0.8385 | 0.8739 | 0.8558 | 0.9460 |
65
- | 0.0009 | 10.0 | 20300 | 0.5594 | 0.8396 | 0.8739 | 0.8564 | 0.9465 |
66
 
67
 
68
  ### Framework versions
 
1
  ---
2
+ base_model: mor40/BulBERT-chitanka-model
3
  tags:
4
  - generated_from_trainer
5
  metrics:
 
17
 
18
  # BulBERT-ner-wikiann
19
 
20
+ This model is a fine-tuned version of [mor40/BulBERT-chitanka-model](https://huggingface.co/mor40/BulBERT-chitanka-model) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.2787
23
+ - Precision: 0.8050
24
+ - Recall: 0.8556
25
+ - F1: 0.8296
26
+ - Accuracy: 0.9446
27
 
28
  ## Model description
29
 
 
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
+ - num_epochs: 5
52
 
53
  ### Training results
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
+ | 0.2436 | 1.0 | 2030 | 0.2391 | 0.7289 | 0.8071 | 0.7660 | 0.9284 |
58
+ | 0.1601 | 2.0 | 4060 | 0.2230 | 0.7698 | 0.8328 | 0.8001 | 0.9380 |
59
+ | 0.102 | 3.0 | 6090 | 0.2441 | 0.7962 | 0.8444 | 0.8196 | 0.9431 |
60
+ | 0.0707 | 4.0 | 8120 | 0.2643 | 0.7998 | 0.8533 | 0.8257 | 0.9444 |
61
+ | 0.0542 | 5.0 | 10150 | 0.2787 | 0.8050 | 0.8556 | 0.8296 | 0.9446 |
 
 
 
 
 
62
 
63
 
64
  ### Framework versions