Hugging Face's logo

stulcrad
/
CNEC_2_0_ext_slavicbert

Token Classification
Transformers
Safetensors
bert
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
stulcrad commited on
Commit
f9d3342
·
verified ·
1 Parent(s): 1962d77

Model save

Browse files
Files changed (2) hide show
  1. README.md +23 -34
  2. model.safetensors +1 -1
README.md CHANGED
@@ -24,16 +24,16 @@ model-index:
24
  metrics:
25
  - name: Precision
26
  type: precision
27
- value: 0.8548621190130624
28
  - name: Recall
29
  type: recall
30
- value: 0.8769230769230769
31
  - name: F1
32
  type: f1
33
- value: 0.8657520823125919
34
  - name: Accuracy
35
  type: accuracy
36
- value: 0.9675938324979914
37
  ---
38
 
39
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -43,11 +43,11 @@ should probably proofread and complete it, then remove this comment. -->
43
 
44
  This model is a fine-tuned version of [DeepPavlov/bert-base-bg-cs-pl-ru-cased](https://huggingface.co/DeepPavlov/bert-base-bg-cs-pl-ru-cased) on the cnec dataset.
45
  It achieves the following results on the evaluation set:
46
- - Loss: 0.2146
47
- - Precision: 0.8549
48
- - Recall: 0.8769
49
- - F1: 0.8658
50
- - Accuracy: 0.9676
51
 
52
  ## Model description
53
 
@@ -67,39 +67,28 @@ More information needed
67
 
68
  The following hyperparameters were used during training:
69
  - learning_rate: 2e-05
70
- - train_batch_size: 16
71
- - eval_batch_size: 16
72
  - seed: 42
73
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
74
  - lr_scheduler_type: linear
75
- - num_epochs: 25
76
 
77
  ### Training results
78
 
79
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
80
  |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
81
- | 0.3662 | 1.12 | 500 | 0.1732 | 0.7518 | 0.8089 | 0.7793 | 0.9512 |
82
- | 0.1571 | 2.24 | 1000 | 0.1441 | 0.8084 | 0.8357 | 0.8219 | 0.9593 |
83
- | 0.116 | 3.36 | 1500 | 0.1440 | 0.8178 | 0.8556 | 0.8363 | 0.9637 |
84
- | 0.0905 | 4.47 | 2000 | 0.1325 | 0.8327 | 0.8794 | 0.8554 | 0.9692 |
85
- | 0.0726 | 5.59 | 2500 | 0.1401 | 0.8410 | 0.8715 | 0.8560 | 0.9677 |
86
- | 0.0595 | 6.71 | 3000 | 0.1536 | 0.8325 | 0.8734 | 0.8525 | 0.9664 |
87
- | 0.0525 | 7.83 | 3500 | 0.1445 | 0.8548 | 0.8710 | 0.8628 | 0.9675 |
88
- | 0.0415 | 8.95 | 4000 | 0.1481 | 0.8513 | 0.8695 | 0.8603 | 0.9677 |
89
- | 0.036 | 10.07 | 4500 | 0.1653 | 0.8491 | 0.8715 | 0.8602 | 0.9684 |
90
- | 0.0304 | 11.19 | 5000 | 0.1709 | 0.8533 | 0.8749 | 0.8640 | 0.9681 |
91
- | 0.0264 | 12.3 | 5500 | 0.1811 | 0.8497 | 0.8779 | 0.8636 | 0.9672 |
92
- | 0.0243 | 13.42 | 6000 | 0.1863 | 0.8460 | 0.8700 | 0.8578 | 0.9659 |
93
- | 0.02 | 14.54 | 6500 | 0.1853 | 0.8511 | 0.8739 | 0.8624 | 0.9677 |
94
- | 0.0183 | 15.66 | 7000 | 0.1952 | 0.8511 | 0.8794 | 0.8650 | 0.9679 |
95
- | 0.0148 | 16.78 | 7500 | 0.2036 | 0.8554 | 0.8804 | 0.8677 | 0.9682 |
96
- | 0.0141 | 17.9 | 8000 | 0.1982 | 0.8528 | 0.8769 | 0.8647 | 0.9680 |
97
- | 0.013 | 19.02 | 8500 | 0.2049 | 0.8562 | 0.8804 | 0.8681 | 0.9685 |
98
- | 0.0116 | 20.13 | 9000 | 0.2044 | 0.8548 | 0.8764 | 0.8655 | 0.9684 |
99
- | 0.0103 | 21.25 | 9500 | 0.2085 | 0.8525 | 0.8809 | 0.8665 | 0.9686 |
100
- | 0.0089 | 22.37 | 10000 | 0.2106 | 0.8529 | 0.8804 | 0.8664 | 0.9681 |
101
- | 0.0091 | 23.49 | 10500 | 0.2113 | 0.8545 | 0.8804 | 0.8673 | 0.9679 |
102
- | 0.0081 | 24.61 | 11000 | 0.2146 | 0.8549 | 0.8769 | 0.8658 | 0.9676 |
103
 
104
 
105
  ### Framework versions
 
24
  metrics:
25
  - name: Precision
26
  type: precision
27
+ value: 0.8578290105667628
28
  - name: Recall
29
  type: recall
30
+ value: 0.8863523573200992
31
  - name: F1
32
  type: f1
33
+ value: 0.8718574566756162
34
  - name: Accuracy
35
  type: accuracy
36
+ value: 0.969659869151012
37
  ---
38
 
39
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
43
 
44
  This model is a fine-tuned version of [DeepPavlov/bert-base-bg-cs-pl-ru-cased](https://huggingface.co/DeepPavlov/bert-base-bg-cs-pl-ru-cased) on the cnec dataset.
45
  It achieves the following results on the evaluation set:
46
+ - Loss: 0.2252
47
+ - Precision: 0.8578
48
+ - Recall: 0.8864
49
+ - F1: 0.8719
50
+ - Accuracy: 0.9697
51
 
52
  ## Model description
53
 
 
67
 
68
  The following hyperparameters were used during training:
69
  - learning_rate: 2e-05
70
+ - train_batch_size: 32
71
+ - eval_batch_size: 32
72
  - seed: 42
73
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
74
  - lr_scheduler_type: linear
75
+ - num_epochs: 50
76
 
77
  ### Training results
78
 
79
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
80
  |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
81
+ | 0.1347 | 4.46 | 1000 | 0.1375 | 0.8279 | 0.8620 | 0.8446 | 0.9656 |
82
+ | 0.0681 | 8.93 | 2000 | 0.1519 | 0.8345 | 0.8710 | 0.8524 | 0.9668 |
83
+ | 0.0406 | 13.39 | 3000 | 0.1663 | 0.8519 | 0.8789 | 0.8652 | 0.9679 |
84
+ | 0.0276 | 17.86 | 4000 | 0.1719 | 0.8623 | 0.8888 | 0.8754 | 0.9690 |
85
+ | 0.02 | 22.32 | 5000 | 0.1920 | 0.8505 | 0.8809 | 0.8654 | 0.9686 |
86
+ | 0.015 | 26.79 | 6000 | 0.1984 | 0.8570 | 0.8893 | 0.8729 | 0.9693 |
87
+ | 0.0108 | 31.25 | 7000 | 0.2048 | 0.8587 | 0.8864 | 0.8723 | 0.9692 |
88
+ | 0.0092 | 35.71 | 8000 | 0.2179 | 0.8606 | 0.8888 | 0.8745 | 0.9696 |
89
+ | 0.0076 | 40.18 | 9000 | 0.2252 | 0.8564 | 0.8878 | 0.8718 | 0.9696 |
90
+ | 0.0057 | 44.64 | 10000 | 0.2262 | 0.8571 | 0.8873 | 0.8720 | 0.9698 |
91
+ | 0.0054 | 49.11 | 11000 | 0.2252 | 0.8578 | 0.8864 | 0.8719 | 0.9697 |
 
 
 
 
 
 
 
 
 
 
 
92
 
93
 
94
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7042c81243751fdbf476c3b7d976f8a5c3dcde031f7a2cf5785539c30265345
3
  size 709120892
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3a632b734c95a1128fcf29b01560d00d01d62f9b54c95dc4ae307d0d5c1d7f4
3
  size 709120892