Hugging Face's logo

stulcrad
/
CNEC_1_1_Supertypes_robeczech-base

Token Classification
Transformers
Safetensors
roberta
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
stulcrad commited on
Commit
0b73d7d
•
1 Parent(s): 6270ff6

Model save

Browse files
Files changed (2) hide show
  1. README.md +42 -27
  2. model.safetensors +1 -1
README.md CHANGED
@@ -25,16 +25,16 @@ model-index:
25
  metrics:
26
  - name: Precision
27
  type: precision
28
- value: 0.8452782462057336
29
  - name: Recall
30
  type: recall
31
- value: 0.8863837312113174
32
  - name: F1
33
  type: f1
34
- value: 0.8653431160984031
35
  - name: Accuracy
36
  type: accuracy
37
- value: 0.9505058365758755
38
  ---
39
 
40
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -44,11 +44,11 @@ should probably proofread and complete it, then remove this comment. -->
44
 
45
  This model is a fine-tuned version of [ufal/robeczech-base](https://huggingface.co/ufal/robeczech-base) on the cnec dataset.
46
  It achieves the following results on the evaluation set:
47
- - Loss: 0.2615
48
- - Precision: 0.8453
49
- - Recall: 0.8864
50
- - F1: 0.8653
51
- - Accuracy: 0.9505
52
 
53
  ## Model description
54
 
@@ -68,8 +68,8 @@ More information needed
68
 
69
  The following hyperparameters were used during training:
70
  - learning_rate: 2e-05
71
- - train_batch_size: 32
72
- - eval_batch_size: 32
73
  - seed: 42
74
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
75
  - lr_scheduler_type: linear
@@ -77,22 +77,37 @@ The following hyperparameters were used during training:
77
 
78
  ### Training results
79
 
80
- | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
81
- |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
82
- | 0.8866 | 3.4 | 500 | 0.3875 | 0.7366 | 0.7467 | 0.7416 | 0.9223 |
83
- | 0.3403 | 6.8 | 1000 | 0.2736 | 0.7700 | 0.8302 | 0.7990 | 0.9337 |
84
- | 0.2206 | 10.2 | 1500 | 0.2203 | 0.8323 | 0.8665 | 0.8490 | 0.9486 |
85
- | 0.1614 | 13.61 | 2000 | 0.2082 | 0.8477 | 0.8811 | 0.8641 | 0.9519 |
86
- | 0.1256 | 17.01 | 2500 | 0.2115 | 0.8401 | 0.8780 | 0.8586 | 0.9507 |
87
- | 0.1008 | 20.41 | 3000 | 0.2311 | 0.8345 | 0.8762 | 0.8549 | 0.9490 |
88
- | 0.0849 | 23.81 | 3500 | 0.2347 | 0.8372 | 0.8868 | 0.8613 | 0.9507 |
89
- | 0.0733 | 27.21 | 4000 | 0.2346 | 0.8425 | 0.8842 | 0.8628 | 0.9508 |
90
- | 0.062 | 30.61 | 4500 | 0.2442 | 0.8440 | 0.8828 | 0.8630 | 0.9500 |
91
- | 0.056 | 34.01 | 5000 | 0.2435 | 0.8449 | 0.8837 | 0.8639 | 0.9512 |
92
- | 0.05 | 37.41 | 5500 | 0.2553 | 0.8375 | 0.8820 | 0.8592 | 0.9497 |
93
- | 0.0456 | 40.82 | 6000 | 0.2575 | 0.8399 | 0.8837 | 0.8613 | 0.9502 |
94
- | 0.0428 | 44.22 | 6500 | 0.2596 | 0.8424 | 0.8886 | 0.8649 | 0.9507 |
95
- | 0.0395 | 47.62 | 7000 | 0.2615 | 0.8453 | 0.8864 | 0.8653 | 0.9505 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
96
 
97
 
98
  ### Framework versions
 
25
  metrics:
26
  - name: Precision
27
  type: precision
28
+ value: 0.844574780058651
29
  - name: Recall
30
  type: recall
31
+ value: 0.8912466843501327
32
  - name: F1
33
  type: f1
34
+ value: 0.8672832867283288
35
  - name: Accuracy
36
  type: accuracy
37
+ value: 0.9517509727626459
38
  ---
39
 
40
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
44
 
45
  This model is a fine-tuned version of [ufal/robeczech-base](https://huggingface.co/ufal/robeczech-base) on the cnec dataset.
46
  It achieves the following results on the evaluation set:
47
+ - Loss: 0.2799
48
+ - Precision: 0.8446
49
+ - Recall: 0.8912
50
+ - F1: 0.8673
51
+ - Accuracy: 0.9518
52
 
53
  ## Model description
54
 
 
68
 
69
  The following hyperparameters were used during training:
70
  - learning_rate: 2e-05
71
+ - train_batch_size: 16
72
+ - eval_batch_size: 16
73
  - seed: 42
74
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
75
  - lr_scheduler_type: linear
 
77
 
78
  ### Training results
79
 
80
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
81
+ |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
82
+ | 1.0614 | 1.7 | 500 | 0.6385 | 0.2880 | 0.1057 | 0.1546 | 0.8234 |
83
+ | 0.5512 | 3.4 | 1000 | 0.3567 | 0.7105 | 0.7542 | 0.7317 | 0.9197 |
84
+ | 0.3472 | 5.1 | 1500 | 0.2644 | 0.7602 | 0.8254 | 0.7914 | 0.9342 |
85
+ | 0.2659 | 6.8 | 2000 | 0.2466 | 0.7945 | 0.8492 | 0.8209 | 0.9389 |
86
+ | 0.2169 | 8.5 | 2500 | 0.2240 | 0.8252 | 0.8621 | 0.8432 | 0.9453 |
87
+ | 0.1797 | 10.2 | 3000 | 0.2113 | 0.8345 | 0.8714 | 0.8525 | 0.9487 |
88
+ | 0.1609 | 11.9 | 3500 | 0.2178 | 0.8213 | 0.8815 | 0.8503 | 0.9487 |
89
+ | 0.1371 | 13.61 | 4000 | 0.2126 | 0.8406 | 0.8811 | 0.8603 | 0.9509 |
90
+ | 0.1237 | 15.31 | 4500 | 0.2127 | 0.8422 | 0.8775 | 0.8595 | 0.9510 |
91
+ | 0.1101 | 17.01 | 5000 | 0.2065 | 0.8520 | 0.8855 | 0.8684 | 0.9538 |
92
+ | 0.0988 | 18.71 | 5500 | 0.2113 | 0.8457 | 0.8895 | 0.8671 | 0.9534 |
93
+ | 0.0904 | 20.41 | 6000 | 0.2280 | 0.8390 | 0.8895 | 0.8635 | 0.9523 |
94
+ | 0.0831 | 22.11 | 6500 | 0.2268 | 0.8430 | 0.8948 | 0.8681 | 0.9532 |
95
+ | 0.0758 | 23.81 | 7000 | 0.2472 | 0.8396 | 0.8864 | 0.8624 | 0.9502 |
96
+ | 0.0713 | 25.51 | 7500 | 0.2377 | 0.8402 | 0.8877 | 0.8633 | 0.9511 |
97
+ | 0.066 | 27.21 | 8000 | 0.2533 | 0.8346 | 0.8855 | 0.8593 | 0.9495 |
98
+ | 0.0591 | 28.91 | 8500 | 0.2449 | 0.8494 | 0.8926 | 0.8704 | 0.9527 |
99
+ | 0.0601 | 30.61 | 9000 | 0.2503 | 0.8421 | 0.8890 | 0.8649 | 0.9527 |
100
+ | 0.0528 | 32.31 | 9500 | 0.2605 | 0.8474 | 0.8935 | 0.8698 | 0.9514 |
101
+ | 0.051 | 34.01 | 10000 | 0.2677 | 0.8389 | 0.8886 | 0.8630 | 0.9511 |
102
+ | 0.0462 | 35.71 | 10500 | 0.2628 | 0.8391 | 0.8921 | 0.8648 | 0.9513 |
103
+ | 0.0438 | 37.41 | 11000 | 0.2629 | 0.8457 | 0.8939 | 0.8691 | 0.9526 |
104
+ | 0.0423 | 39.12 | 11500 | 0.2673 | 0.8406 | 0.8930 | 0.8660 | 0.9502 |
105
+ | 0.0395 | 40.82 | 12000 | 0.2700 | 0.8423 | 0.8904 | 0.8657 | 0.9518 |
106
+ | 0.0386 | 42.52 | 12500 | 0.2716 | 0.8486 | 0.8943 | 0.8709 | 0.9528 |
107
+ | 0.0384 | 44.22 | 13000 | 0.2727 | 0.8465 | 0.8921 | 0.8687 | 0.9523 |
108
+ | 0.0352 | 45.92 | 13500 | 0.2741 | 0.8494 | 0.8926 | 0.8704 | 0.9526 |
109
+ | 0.0351 | 47.62 | 14000 | 0.2776 | 0.8469 | 0.8926 | 0.8691 | 0.9520 |
110
+ | 0.0327 | 49.32 | 14500 | 0.2799 | 0.8446 | 0.8912 | 0.8673 | 0.9518 |
111
 
112
 
113
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d833ee8cade3aa302b99121d5dec8fabd1a18f0332a23321d89a05bb55319932
3
  size 501629388
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:222876d40e5f6da975945537284083cbe5fea75dc63bc6b31dbf314f45eb5da1
3
  size 501629388