Hugging Face's logo

stulcrad
/
CNEC2_0_Supertypes_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
dfa0a48
•
1 Parent(s): a48187e

Model save

Browse files
Files changed (2) hide show
  1. README.md +20 -32
  2. runs/Mar07_15-53-13_g01/events.out.tfevents.1709823194.g01.750675.0 +2 -2
README.md CHANGED
@@ -25,16 +25,16 @@ model-index:
25
  metrics:
26
  - name: Precision
27
  type: precision
28
- value: 0.8214447978191731
29
  - name: Recall
30
  type: recall
31
- value: 0.8725868725868726
32
  - name: F1
33
  type: f1
34
- value: 0.8462438567750995
35
  - name: Accuracy
36
  type: accuracy
37
- value: 0.9689700130378096
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: 0.1759
48
- - Precision: 0.8214
49
- - Recall: 0.8726
50
- - F1: 0.8462
51
- - Accuracy: 0.9690
52
 
53
  ## Model description
54
 
@@ -67,38 +67,26 @@ More information needed
67
  ### Training hyperparameters
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
  - lr_scheduler_warmup_ratio: 0.1
77
- - lr_scheduler_warmup_steps: 1000
78
- - num_epochs: 10
79
 
80
  ### Training results
81
 
82
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
83
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
84
- | 0.9224 | 0.56 | 500 | 0.2309 | 0.5594 | 0.6863 | 0.6164 | 0.9358 |
85
- | 0.2449 | 1.11 | 1000 | 0.1960 | 0.6745 | 0.8142 | 0.7378 | 0.9525 |
86
- | 0.204 | 1.67 | 1500 | 0.1701 | 0.7256 | 0.8079 | 0.7646 | 0.9571 |
87
- | 0.1694 | 2.22 | 2000 | 0.1526 | 0.7605 | 0.8567 | 0.8057 | 0.9640 |
88
- | 0.1392 | 2.78 | 2500 | 0.1607 | 0.7697 | 0.8485 | 0.8072 | 0.9620 |
89
- | 0.1191 | 3.33 | 3000 | 0.1528 | 0.7969 | 0.8596 | 0.8270 | 0.9646 |
90
- | 0.1128 | 3.89 | 3500 | 0.1552 | 0.7668 | 0.8711 | 0.8156 | 0.9610 |
91
- | 0.095 | 4.44 | 4000 | 0.1678 | 0.7658 | 0.8615 | 0.8108 | 0.9632 |
92
- | 0.0979 | 5.0 | 4500 | 0.1432 | 0.8079 | 0.8625 | 0.8343 | 0.9672 |
93
- | 0.0764 | 5.56 | 5000 | 0.1548 | 0.8098 | 0.8528 | 0.8307 | 0.9671 |
94
- | 0.0829 | 6.11 | 5500 | 0.1423 | 0.8128 | 0.8653 | 0.8382 | 0.9672 |
95
- | 0.0648 | 6.67 | 6000 | 0.1548 | 0.8038 | 0.8760 | 0.8383 | 0.9673 |
96
- | 0.0529 | 7.22 | 6500 | 0.1653 | 0.8139 | 0.8716 | 0.8418 | 0.9675 |
97
- | 0.0483 | 7.78 | 7000 | 0.1630 | 0.8186 | 0.8649 | 0.8411 | 0.9680 |
98
- | 0.0494 | 8.33 | 7500 | 0.1709 | 0.8233 | 0.8682 | 0.8452 | 0.9686 |
99
- | 0.0389 | 8.89 | 8000 | 0.1757 | 0.8211 | 0.8726 | 0.8460 | 0.9687 |
100
- | 0.0356 | 9.44 | 8500 | 0.1740 | 0.8242 | 0.8736 | 0.8482 | 0.9692 |
101
- | 0.0337 | 10.0 | 9000 | 0.1759 | 0.8214 | 0.8726 | 0.8462 | 0.9690 |
102
 
103
 
104
  ### Framework versions
 
25
  metrics:
26
  - name: Precision
27
  type: precision
28
+ value: 0.8222823635543527
29
  - name: Recall
30
  type: recall
31
+ value: 0.8798262548262549
32
  - name: F1
33
  type: f1
34
+ value: 0.8500816041035206
35
  - name: Accuracy
36
  type: accuracy
37
+ value: 0.9681297986382732
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: 0.2071
48
+ - Precision: 0.8223
49
+ - Recall: 0.8798
50
+ - F1: 0.8501
51
+ - Accuracy: 0.9681
52
 
53
  ## Model description
54
 
 
67
  ### Training hyperparameters
68
 
69
  The following hyperparameters were used during training:
70
+ - learning_rate: 5e-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
  - lr_scheduler_warmup_ratio: 0.1
77
+ - lr_scheduler_warmup_steps: 500
78
+ - num_epochs: 15
79
 
80
  ### Training results
81
 
82
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
83
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
84
+ | 0.4978 | 2.22 | 500 | 0.1737 | 0.6882 | 0.8118 | 0.7449 | 0.9548 |
85
+ | 0.149 | 4.44 | 1000 | 0.1573 | 0.7540 | 0.8552 | 0.8014 | 0.9596 |
86
+ | 0.0796 | 6.67 | 1500 | 0.1530 | 0.8024 | 0.8760 | 0.8376 | 0.9648 |
87
+ | 0.0473 | 8.89 | 2000 | 0.1539 | 0.8051 | 0.8731 | 0.8377 | 0.9675 |
88
+ | 0.0272 | 11.11 | 2500 | 0.2028 | 0.7973 | 0.8581 | 0.8266 | 0.9643 |
89
+ | 0.0154 | 13.33 | 3000 | 0.2071 | 0.8223 | 0.8798 | 0.8501 | 0.9681 |
 
 
 
 
 
 
 
 
 
 
 
 
90
 
91
 
92
  ### Framework versions
runs/Mar07_15-53-13_g01/events.out.tfevents.1709823194.g01.750675.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b6ca1dac5889e2e851ce0ca78d979a9d44c93f066bcc358d471b1572a4c468e8
3
- size 8756
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4e6ab5b4c9826180a8dc9dc591df5b3ceb5ff1ee91dc0eaadd451cd5ca6e4dc
3
+ size 9110