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
4eca7d5
•
1 Parent(s): f68bc61

Model save

Browse files
Files changed (1) hide show
  1. README.md +21 -34
README.md CHANGED
@@ -25,16 +25,16 @@ model-index:
25
  metrics:
26
  - name: Precision
27
  type: precision
28
- value: 0.8427382053654024
29
  - name: Recall
30
  type: recall
31
- value: 0.8793436293436293
32
  - name: F1
33
  type: f1
34
- value: 0.8606518658478979
35
  - name: Accuracy
36
  type: accuracy
37
- value: 0.9671736925974214
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: 0.2674
48
- - Precision: 0.8427
49
- - Recall: 0.8793
50
- - F1: 0.8607
51
- - Accuracy: 0.9672
52
 
53
  ## Model description
54
 
@@ -73,34 +73,21 @@ The following hyperparameters were used during training:
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
- | 0.5221 | 1.11 | 500 | 0.1718 | 0.6648 | 0.8012 | 0.7266 | 0.9535 |
83
- | 0.1777 | 2.22 | 1000 | 0.1397 | 0.7499 | 0.8393 | 0.7921 | 0.9627 |
84
- | 0.1321 | 3.33 | 1500 | 0.1383 | 0.7760 | 0.8711 | 0.8208 | 0.9655 |
85
- | 0.1132 | 4.44 | 2000 | 0.1456 | 0.7646 | 0.8542 | 0.8069 | 0.9636 |
86
- | 0.1008 | 5.56 | 2500 | 0.1442 | 0.7750 | 0.8692 | 0.8194 | 0.9648 |
87
- | 0.0782 | 6.67 | 3000 | 0.1516 | 0.8107 | 0.8663 | 0.8376 | 0.9657 |
88
- | 0.0692 | 7.78 | 3500 | 0.1690 | 0.8023 | 0.8620 | 0.8311 | 0.9660 |
89
- | 0.0582 | 8.89 | 4000 | 0.1591 | 0.8125 | 0.8847 | 0.8470 | 0.9672 |
90
- | 0.0511 | 10.0 | 4500 | 0.1813 | 0.8033 | 0.8832 | 0.8414 | 0.9661 |
91
- | 0.0432 | 11.11 | 5000 | 0.1833 | 0.8231 | 0.8822 | 0.8516 | 0.9669 |
92
- | 0.0381 | 12.22 | 5500 | 0.2097 | 0.8062 | 0.8634 | 0.8338 | 0.9659 |
93
- | 0.0328 | 13.33 | 6000 | 0.2043 | 0.8026 | 0.8711 | 0.8355 | 0.9661 |
94
- | 0.0292 | 14.44 | 6500 | 0.2217 | 0.8255 | 0.8769 | 0.8505 | 0.9669 |
95
- | 0.0247 | 15.56 | 7000 | 0.2411 | 0.8297 | 0.8745 | 0.8515 | 0.9667 |
96
- | 0.0206 | 16.67 | 7500 | 0.2425 | 0.8255 | 0.8764 | 0.8502 | 0.9663 |
97
- | 0.0184 | 17.78 | 8000 | 0.2405 | 0.8329 | 0.8586 | 0.8455 | 0.9668 |
98
- | 0.0157 | 18.89 | 8500 | 0.2521 | 0.8314 | 0.8832 | 0.8565 | 0.9677 |
99
- | 0.0134 | 20.0 | 9000 | 0.2504 | 0.8349 | 0.8764 | 0.8552 | 0.9671 |
100
- | 0.0116 | 21.11 | 9500 | 0.2570 | 0.8344 | 0.8779 | 0.8556 | 0.9678 |
101
- | 0.0109 | 22.22 | 10000 | 0.2570 | 0.8320 | 0.8793 | 0.8550 | 0.9677 |
102
- | 0.0093 | 23.33 | 10500 | 0.2639 | 0.8373 | 0.8793 | 0.8578 | 0.9674 |
103
- | 0.0086 | 24.44 | 11000 | 0.2674 | 0.8427 | 0.8793 | 0.8607 | 0.9672 |
104
 
105
 
106
  ### Framework versions
 
25
  metrics:
26
  - name: Precision
27
  type: precision
28
+ value: 0.8280399274047187
29
  - name: Recall
30
  type: recall
31
+ value: 0.8807915057915058
32
  - name: F1
33
  type: f1
34
+ value: 0.8536014967259121
35
  - name: Accuracy
36
  type: accuracy
37
+ value: 0.9694915254237289
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: 0.1664
48
+ - Precision: 0.8280
49
+ - Recall: 0.8808
50
+ - F1: 0.8536
51
+ - Accuracy: 0.9695
52
 
53
  ## Model description
54
 
 
73
  - seed: 42
74
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
75
  - lr_scheduler_type: linear
76
+ - num_epochs: 10
77
 
78
  ### Training results
79
 
80
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
81
+ |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
82
+ | 0.3183 | 1.11 | 500 | 0.1528 | 0.6999 | 0.7934 | 0.7437 | 0.9583 |
83
+ | 0.144 | 2.22 | 1000 | 0.1302 | 0.7521 | 0.8639 | 0.8041 | 0.9648 |
84
+ | 0.1015 | 3.33 | 1500 | 0.1431 | 0.8003 | 0.8721 | 0.8346 | 0.9678 |
85
+ | 0.0807 | 4.44 | 2000 | 0.1355 | 0.7840 | 0.8740 | 0.8266 | 0.9680 |
86
+ | 0.066 | 5.56 | 2500 | 0.1413 | 0.8196 | 0.8793 | 0.8484 | 0.9691 |
87
+ | 0.0492 | 6.67 | 3000 | 0.1461 | 0.8132 | 0.8803 | 0.8454 | 0.9700 |
88
+ | 0.0401 | 7.78 | 3500 | 0.1577 | 0.8229 | 0.8769 | 0.8491 | 0.9690 |
89
+ | 0.0312 | 8.89 | 4000 | 0.1637 | 0.8242 | 0.8822 | 0.8522 | 0.9700 |
90
+ | 0.0265 | 10.0 | 4500 | 0.1664 | 0.8280 | 0.8808 | 0.8536 | 0.9695 |
 
 
 
 
 
 
 
 
 
 
 
 
 
91
 
92
 
93
  ### Framework versions