Hugging Face's logo

stulcrad
/
Robeczech-CERED2

Transformers
TensorBoard
Safetensors
roberta
Generated from Trainer
Model card Files
xet
Metrics Community
stulcrad commited on
Commit
d54ae6e
·
verified ·
1 Parent(s): fd73d1d

End of training

Browse files
Files changed (5) hide show
  1. README.md +33 -28
  2. model.safetensors +1 -1
  3. runs/Mar31_10-43-37_dgx10/events.out.tfevents.1743410625.dgx10.924525.0 +3 -0
  4. runs/Mar31_10-43-37_dgx10/events.out.tfevents.1743425242.dgx10.924525.1 +3 -0
  5. training_args.bin +2 -2
README.md CHANGED
@@ -5,29 +5,29 @@ base_model: ufal/robeczech-base
5
  tags:
6
  - generated_from_trainer
7
  datasets:
8
- - stulcrad/CERED-2
9
  metrics:
10
  - accuracy
11
  model-index:
12
- - name: Robeczech-2
13
  results: []
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
- # Robeczech-2
20
 
21
- This model is a fine-tuned version of [ufal/robeczech-base](https://huggingface.co/ufal/robeczech-base) on the Czech Relationship Extraction Dataset (http://hdl.handle.net/11234/1-3265).
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.8211
24
- - Accuracy: 0.8893
25
- - Micro Precision: 0.8893
26
- - Micro Recall: 0.8893
27
- - Micro F1: 0.8893
28
- - Macro Precision: 0.8687
29
- - Macro Recall: 0.8432
30
- - Macro F1: 0.8512
31
 
32
  ## Model description
33
 
@@ -46,26 +46,31 @@ More information needed
46
  ### Training hyperparameters
47
 
48
  The following hyperparameters were used during training:
49
- - learning_rate: 5e-05
50
- - train_batch_size: 12
51
- - eval_batch_size: 12
52
  - seed: 42
 
 
53
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
54
- - lr_scheduler_type: linear
55
- - num_epochs: 8
 
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | Micro Precision | Micro Recall | Micro F1 | Macro Precision | Macro Recall | Macro F1 |
60
- |:-------------:|:-----:|:------:|:---------------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:--------:|
61
- | 0.5345 | 1.0 | 30148 | 0.5007 | 0.8546 | 0.8546 | 0.8546 | 0.8546 | 0.8145 | 0.7847 | 0.7893 |
62
- | 0.4516 | 2.0 | 60296 | 0.4827 | 0.8759 | 0.8759 | 0.8759 | 0.8759 | 0.8478 | 0.8281 | 0.8293 |
63
- | 0.3227 | 3.0 | 90444 | 0.4958 | 0.8839 | 0.8839 | 0.8839 | 0.8839 | 0.8564 | 0.8298 | 0.8357 |
64
- | 0.2641 | 4.0 | 120592 | 0.5288 | 0.8850 | 0.8850 | 0.8850 | 0.8850 | 0.8506 | 0.8405 | 0.8411 |
65
- | 0.2054 | 5.0 | 150740 | 0.6073 | 0.8862 | 0.8862 | 0.8862 | 0.8862 | 0.8487 | 0.8435 | 0.8419 |
66
- | 0.1454 | 6.0 | 180888 | 0.6945 | 0.8862 | 0.8862 | 0.8862 | 0.8862 | 0.8545 | 0.8399 | 0.8415 |
67
- | 0.0747 | 7.0 | 211036 | 0.7675 | 0.8926 | 0.8926 | 0.8926 | 0.8926 | 0.8584 | 0.8525 | 0.8516 |
68
- | 0.0472 | 8.0 | 241184 | 0.8169 | 0.8917 | 0.8917 | 0.8917 | 0.8917 | 0.8561 | 0.8471 | 0.8477 |
 
 
69
 
70
 
71
  ### Framework versions
@@ -73,4 +78,4 @@ The following hyperparameters were used during training:
73
  - Transformers 4.46.2
74
  - Pytorch 2.5.1+cu124
75
  - Datasets 3.1.0
76
- - Tokenizers 0.20.3
 
5
  tags:
6
  - generated_from_trainer
7
  datasets:
8
+ - generator
9
  metrics:
10
  - accuracy
11
  model-index:
12
+ - name: Robeczech-CERED2
13
  results: []
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
+ # Robeczech-CERED2
20
 
21
+ This model is a fine-tuned version of [ufal/robeczech-base](https://huggingface.co/ufal/robeczech-base) on the generator dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.5511
24
+ - Accuracy: 0.8942
25
+ - Micro Precision: 0.8942
26
+ - Micro Recall: 0.8942
27
+ - Micro F1: 0.8942
28
+ - Macro Precision: 0.8789
29
+ - Macro Recall: 0.8525
30
+ - Macro F1: 0.8597
31
 
32
  ## Model description
33
 
 
46
  ### Training hyperparameters
47
 
48
  The following hyperparameters were used during training:
49
+ - learning_rate: 1e-05
50
+ - train_batch_size: 16
51
+ - eval_batch_size: 16
52
  - seed: 42
53
+ - gradient_accumulation_steps: 2
54
+ - total_train_batch_size: 32
55
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
56
+ - lr_scheduler_type: reduce_lr_on_plateau
57
+ - lr_scheduler_warmup_steps: 1000
58
+ - num_epochs: 10
59
 
60
  ### Training results
61
 
62
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Micro Precision | Micro Recall | Micro F1 | Macro Precision | Macro Recall | Macro F1 |
63
+ |:-------------:|:------:|:------:|:---------------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:--------:|
64
+ | 0.5022 | 1.0000 | 11305 | 0.4532 | 0.8668 | 0.8668 | 0.8668 | 0.8668 | 0.8227 | 0.7921 | 0.7968 |
65
+ | 0.3915 | 2.0 | 22611 | 0.3896 | 0.8853 | 0.8853 | 0.8853 | 0.8853 | 0.8454 | 0.8386 | 0.8359 |
66
+ | 0.2895 | 3.0000 | 33916 | 0.3825 | 0.8929 | 0.8929 | 0.8929 | 0.8929 | 0.8591 | 0.8409 | 0.8448 |
67
+ | 0.2404 | 4.0 | 45222 | 0.4012 | 0.8944 | 0.8944 | 0.8944 | 0.8944 | 0.8545 | 0.8517 | 0.8489 |
68
+ | 0.1901 | 5.0000 | 56527 | 0.4284 | 0.8988 | 0.8988 | 0.8988 | 0.8988 | 0.8640 | 0.8599 | 0.8577 |
69
+ | 0.1586 | 6.0 | 67833 | 0.4548 | 0.8983 | 0.8983 | 0.8983 | 0.8983 | 0.8518 | 0.8607 | 0.8526 |
70
+ | 0.1325 | 7.0000 | 79138 | 0.4821 | 0.9020 | 0.9020 | 0.9020 | 0.9020 | 0.8729 | 0.8619 | 0.8641 |
71
+ | 0.1124 | 8.0 | 90444 | 0.5333 | 0.8976 | 0.8976 | 0.8976 | 0.8976 | 0.8685 | 0.8518 | 0.8556 |
72
+ | 0.0883 | 9.0000 | 101749 | 0.6051 | 0.8977 | 0.8977 | 0.8977 | 0.8977 | 0.8615 | 0.8558 | 0.8551 |
73
+ | 0.0874 | 9.9996 | 113050 | 0.6492 | 0.9006 | 0.9006 | 0.9006 | 0.9006 | 0.8601 | 0.8637 | 0.8581 |
74
 
75
 
76
  ### Framework versions
 
78
  - Transformers 4.46.2
79
  - Pytorch 2.5.1+cu124
80
  - Datasets 3.1.0
81
+ - Tokenizers 0.20.3
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:484e955c30c46038a98f82fa40ca8c9ecc3e9e9bbfe612daf67f1963c3ba6b69
3
  size 504532408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce4b09c7a91e77f1ec8f13d478d5061116a62fd587618a3639544a77d651184c
3
  size 504532408
runs/Mar31_10-43-37_dgx10/events.out.tfevents.1743410625.dgx10.924525.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b52dba5252020f27c59ad176a33033cee96d66a03abaafbacbd9a59764ac8330
3
+ size 65381
runs/Mar31_10-43-37_dgx10/events.out.tfevents.1743425242.dgx10.924525.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb61f9f500d4c713bb0287de5c8fd9e76a5a2f18334bef07c44b189aedc46894
3
+ size 757
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d3e17b494324463cb1a546949c9b230fa26d849c55fd88e89d57394c1a5e5e9a
3
- size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd347c115947f325102151d0f0f28f4b3955219062b7a95ebb6e34f82f96a83f
3
+ size 5368