Hugging Face's logo

stulcrad
/
CNEC1_1_extended_xlm-roberta-large

Token Classification
Transformers
TensorBoard
Safetensors
xlm-roberta
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
stulcrad commited on
Commit
261e469
•
1 Parent(s): 77f49b7

End of training

Browse files
Files changed (5) hide show
  1. README.md +25 -19
  2. config.json +2 -2
  3. model.safetensors +1 -1
  4. tokenizer_config.json +1 -0
  5. training_args.bin +1 -1
README.md CHANGED
@@ -25,16 +25,16 @@ model-index:
25
  metrics:
26
  - name: Precision
27
  type: precision
28
- value: 0.843939393939394
29
  - name: Recall
30
  type: recall
31
- value: 0.8767051416579223
32
  - name: F1
33
  type: f1
34
- value: 0.8600102933607823
35
  - name: Accuracy
36
  type: accuracy
37
- value: 0.9590540063536886
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 [FacebookAI/xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) on the cnec dataset.
46
  It achieves the following results on the evaluation set:
47
- - Loss: 0.2330
48
- - Precision: 0.8439
49
- - Recall: 0.8767
50
- - F1: 0.8600
51
- - Accuracy: 0.9591
52
 
53
  ## Model description
54
 
@@ -68,24 +68,30 @@ 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
76
- - num_epochs: 25
77
 
78
  ### Training results
79
 
80
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
81
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
82
- | 0.6037 | 3.42 | 500 | 0.2144 | 0.6741 | 0.7263 | 0.6993 | 0.9407 |
83
- | 0.2517 | 6.85 | 1000 | 0.1774 | 0.7915 | 0.8381 | 0.8141 | 0.9575 |
84
- | 0.1854 | 10.27 | 1500 | 0.1633 | 0.8251 | 0.8621 | 0.8432 | 0.9632 |
85
- | 0.1476 | 13.7 | 2000 | 0.1610 | 0.8356 | 0.8803 | 0.8574 | 0.9664 |
86
- | 0.1248 | 17.12 | 2500 | 0.1657 | 0.8459 | 0.8830 | 0.8640 | 0.9682 |
87
- | 0.1086 | 20.55 | 3000 | 0.1671 | 0.8542 | 0.8862 | 0.8699 | 0.9683 |
88
- | 0.0985 | 23.97 | 3500 | 0.1610 | 0.8600 | 0.8931 | 0.8762 | 0.9700 |
 
 
 
 
 
 
89
 
90
 
91
  ### Framework versions
 
25
  metrics:
26
  - name: Precision
27
  type: precision
28
+ value: 0.8567735263702172
29
  - name: Recall
30
  type: recall
31
+ value: 0.8856226616782469
32
  - name: F1
33
  type: f1
34
+ value: 0.8709592641261498
35
  - name: Accuracy
36
  type: accuracy
37
+ value: 0.9657602026679334
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 [FacebookAI/xlm-roberta-large](https://huggingface.co/FacebookAI/xlm-roberta-large) on the cnec dataset.
46
  It achieves the following results on the evaluation set:
47
+ - Loss: 0.2099
48
+ - Precision: 0.8568
49
+ - Recall: 0.8856
50
+ - F1: 0.8710
51
+ - Accuracy: 0.9658
52
 
53
  ## Model description
54
 
 
68
 
69
  The following hyperparameters were used during training:
70
  - learning_rate: 2e-05
71
+ - train_batch_size: 8
72
+ - eval_batch_size: 8
73
  - seed: 42
74
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
75
  - lr_scheduler_type: linear
76
+ - num_epochs: 12
77
 
78
  ### Training results
79
 
80
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
81
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
82
+ | 0.4598 | 0.86 | 500 | 0.1804 | 0.7750 | 0.8156 | 0.7948 | 0.9556 |
83
+ | 0.2213 | 1.72 | 1000 | 0.1511 | 0.8054 | 0.8338 | 0.8193 | 0.9623 |
84
+ | 0.1614 | 2.58 | 1500 | 0.1517 | 0.8252 | 0.8632 | 0.8438 | 0.9633 |
85
+ | 0.1436 | 3.44 | 2000 | 0.1692 | 0.8207 | 0.8541 | 0.8371 | 0.9619 |
86
+ | 0.1199 | 4.3 | 2500 | 0.1638 | 0.8309 | 0.8589 | 0.8447 | 0.9626 |
87
+ | 0.1076 | 5.16 | 3000 | 0.1710 | 0.8423 | 0.8819 | 0.8616 | 0.9659 |
88
+ | 0.0907 | 6.02 | 3500 | 0.1640 | 0.8366 | 0.8594 | 0.8479 | 0.9651 |
89
+ | 0.0729 | 6.88 | 4000 | 0.1657 | 0.8411 | 0.8658 | 0.8533 | 0.9651 |
90
+ | 0.0678 | 7.75 | 4500 | 0.1919 | 0.8444 | 0.8728 | 0.8583 | 0.9641 |
91
+ | 0.0577 | 8.61 | 5000 | 0.1877 | 0.8429 | 0.8744 | 0.8583 | 0.9644 |
92
+ | 0.0495 | 9.47 | 5500 | 0.1921 | 0.8512 | 0.8803 | 0.8655 | 0.9652 |
93
+ | 0.0462 | 10.33 | 6000 | 0.1952 | 0.8509 | 0.8787 | 0.8646 | 0.9658 |
94
+ | 0.0383 | 11.19 | 6500 | 0.2099 | 0.8568 | 0.8856 | 0.8710 | 0.9658 |
95
 
96
 
97
  ### Framework versions
config.json CHANGED
@@ -3,12 +3,12 @@
3
  "architectures": [
4
  "XLMRobertaForTokenClassification"
5
  ],
6
- "attention_probs_dropout_prob": 0.2,
7
  "bos_token_id": 0,
8
  "classifier_dropout": null,
9
  "eos_token_id": 2,
10
  "hidden_act": "gelu",
11
- "hidden_dropout_prob": 0.4,
12
  "hidden_size": 1024,
13
  "id2label": {
14
  "0": "O",
 
3
  "architectures": [
4
  "XLMRobertaForTokenClassification"
5
  ],
6
+ "attention_probs_dropout_prob": 0.25,
7
  "bos_token_id": 0,
8
  "classifier_dropout": null,
9
  "eos_token_id": 2,
10
  "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.25,
12
  "hidden_size": 1024,
13
  "id2label": {
14
  "0": "O",
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:59bd35b85e3c1ea53167b581e47c0071bf767d9c2a95d08524089d7cc6eb654b
3
  size 2235473356
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b11bd00e1954f7ed11c669792e2d81b588166a305d4b4712ba29b12944d69658
3
  size 2235473356
tokenizer_config.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "added_tokens_decoder": {
3
  "0": {
4
  "content": "<s>",
 
1
  {
2
+ "add_prefix_space": true,
3
  "added_tokens_decoder": {
4
  "0": {
5
  "content": "<s>",
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8d950f5cbacd47c735cf795a7bb9260f9ed5e8763601f8428d23f4d02bd12652
3
  size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f673b0ea2252981f0f4d60f8b28f7976a7e2c8590676c86ae69d8fec6a5ac560
3
  size 4728