Hugging Face's logo

emilys
/
twitter-roberta-base-CoNLL

Token Classification
Transformers
PyTorch
roberta
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
2
emilys commited on
Commit
06e92cd
·
1 Parent(s): 972806d

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +31 -31
README.md CHANGED
@@ -21,16 +21,16 @@ model-index:
21
  metrics:
22
  - name: Precision
23
  type: precision
24
- value: 0.7243867243867244
25
  - name: Recall
26
  type: recall
27
- value: 0.6004784688995215
28
  - name: F1
29
  type: f1
30
- value: 0.6566383257030739
31
  - name: Accuracy
32
  type: accuracy
33
- value: 0.9638339795334647
34
  ---
35
 
36
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -40,11 +40,11 @@ should probably proofread and complete it, then remove this comment. -->
40
 
41
  This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base](https://huggingface.co/cardiffnlp/twitter-roberta-base) on the wnut_17 dataset.
42
  It achieves the following results on the evaluation set:
43
- - Loss: 0.2139
44
- - Precision: 0.7244
45
- - Recall: 0.6005
46
- - F1: 0.6566
47
- - Accuracy: 0.9638
48
 
49
  ## Model description
50
 
@@ -63,7 +63,7 @@ More information needed
63
  ### Training hyperparameters
64
 
65
  The following hyperparameters were used during training:
66
- - learning_rate: 5e-05
67
  - train_batch_size: 64
68
  - eval_batch_size: 1024
69
  - seed: 42
@@ -75,27 +75,27 @@ The following hyperparameters were used during training:
75
 
76
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
77
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
78
- | No log | 0.46 | 25 | 0.2883 | 0.2361 | 0.0203 | 0.0374 | 0.9230 |
79
- | No log | 0.93 | 50 | 0.2093 | 0.7016 | 0.4330 | 0.5355 | 0.9475 |
80
- | No log | 1.39 | 75 | 0.1807 | 0.6915 | 0.5766 | 0.6288 | 0.9579 |
81
- | No log | 1.85 | 100 | 0.1702 | 0.6573 | 0.5873 | 0.6203 | 0.9608 |
82
- | No log | 2.31 | 125 | 0.1767 | 0.7207 | 0.5586 | 0.6294 | 0.9616 |
83
- | No log | 2.78 | 150 | 0.1737 | 0.7136 | 0.5754 | 0.6371 | 0.9617 |
84
- | No log | 3.24 | 175 | 0.1733 | 0.7247 | 0.6077 | 0.6610 | 0.9635 |
85
- | No log | 3.7 | 200 | 0.1683 | 0.6997 | 0.6328 | 0.6646 | 0.9645 |
86
- | No log | 4.17 | 225 | 0.1790 | 0.7162 | 0.6280 | 0.6692 | 0.9648 |
87
- | No log | 4.63 | 250 | 0.1847 | 0.7283 | 0.6029 | 0.6597 | 0.9645 |
88
- | No log | 5.09 | 275 | 0.1949 | 0.7248 | 0.6112 | 0.6632 | 0.9638 |
89
- | No log | 5.56 | 300 | 0.1853 | 0.7127 | 0.6232 | 0.6650 | 0.9648 |
90
- | No log | 6.02 | 325 | 0.1807 | 0.7026 | 0.6388 | 0.6692 | 0.9645 |
91
- | No log | 6.48 | 350 | 0.2000 | 0.7191 | 0.6184 | 0.6650 | 0.9642 |
92
- | No log | 6.94 | 375 | 0.1997 | 0.7191 | 0.6124 | 0.6615 | 0.9638 |
93
- | No log | 7.41 | 400 | 0.2035 | 0.7193 | 0.6160 | 0.6637 | 0.9640 |
94
- | No log | 7.87 | 425 | 0.1977 | 0.6948 | 0.6208 | 0.6557 | 0.9643 |
95
- | No log | 8.33 | 450 | 0.2048 | 0.7174 | 0.6256 | 0.6684 | 0.9645 |
96
- | No log | 8.8 | 475 | 0.2073 | 0.7197 | 0.6112 | 0.6611 | 0.9642 |
97
- | 0.07 | 9.26 | 500 | 0.2139 | 0.7244 | 0.6005 | 0.6566 | 0.9638 |
98
- | 0.07 | 9.72 | 525 | 0.2049 | 0.7058 | 0.6256 | 0.6633 | 0.9647 |
99
 
100
 
101
  ### Framework versions
 
21
  metrics:
22
  - name: Precision
23
  type: precision
24
+ value: 0.7024901703800787
25
  - name: Recall
26
  type: recall
27
+ value: 0.6411483253588517
28
  - name: F1
29
  type: f1
30
+ value: 0.6704190118824266
31
  - name: Accuracy
32
  type: accuracy
33
+ value: 0.9645967075573635
34
  ---
35
 
36
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
40
 
41
  This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base](https://huggingface.co/cardiffnlp/twitter-roberta-base) on the wnut_17 dataset.
42
  It achieves the following results on the evaluation set:
43
+ - Loss: 0.1880
44
+ - Precision: 0.7025
45
+ - Recall: 0.6411
46
+ - F1: 0.6704
47
+ - Accuracy: 0.9646
48
 
49
  ## Model description
50
 
 
63
  ### Training hyperparameters
64
 
65
  The following hyperparameters were used during training:
66
+ - learning_rate: 2e-05
67
  - train_batch_size: 64
68
  - eval_batch_size: 1024
69
  - seed: 42
 
75
 
76
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
77
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
78
+ | No log | 0.46 | 25 | 0.3912 | 0.0 | 0.0 | 0.0 | 0.9205 |
79
+ | No log | 0.93 | 50 | 0.2847 | 0.25 | 0.0024 | 0.0047 | 0.9209 |
80
+ | No log | 1.39 | 75 | 0.2449 | 0.5451 | 0.3469 | 0.4240 | 0.9426 |
81
+ | No log | 1.85 | 100 | 0.1946 | 0.6517 | 0.4856 | 0.5565 | 0.9492 |
82
+ | No log | 2.31 | 125 | 0.1851 | 0.6921 | 0.5646 | 0.6219 | 0.9581 |
83
+ | No log | 2.78 | 150 | 0.1672 | 0.6867 | 0.5873 | 0.6331 | 0.9594 |
84
+ | No log | 3.24 | 175 | 0.1675 | 0.6787 | 0.5837 | 0.6277 | 0.9615 |
85
+ | No log | 3.7 | 200 | 0.1644 | 0.6765 | 0.6328 | 0.6539 | 0.9638 |
86
+ | No log | 4.17 | 225 | 0.1672 | 0.6997 | 0.6495 | 0.6737 | 0.9640 |
87
+ | No log | 4.63 | 250 | 0.1652 | 0.6915 | 0.6435 | 0.6667 | 0.9649 |
88
+ | No log | 5.09 | 275 | 0.1882 | 0.7067 | 0.6053 | 0.6521 | 0.9629 |
89
+ | No log | 5.56 | 300 | 0.1783 | 0.7128 | 0.6352 | 0.6717 | 0.9645 |
90
+ | No log | 6.02 | 325 | 0.1813 | 0.7011 | 0.6172 | 0.6565 | 0.9639 |
91
+ | No log | 6.48 | 350 | 0.1804 | 0.7139 | 0.6447 | 0.6776 | 0.9647 |
92
+ | No log | 6.94 | 375 | 0.1902 | 0.7218 | 0.6268 | 0.6709 | 0.9641 |
93
+ | No log | 7.41 | 400 | 0.1883 | 0.7106 | 0.6316 | 0.6688 | 0.9641 |
94
+ | No log | 7.87 | 425 | 0.1862 | 0.7067 | 0.6340 | 0.6683 | 0.9643 |
95
+ | No log | 8.33 | 450 | 0.1882 | 0.7053 | 0.6328 | 0.6671 | 0.9639 |
96
+ | No log | 8.8 | 475 | 0.1919 | 0.7055 | 0.6304 | 0.6658 | 0.9638 |
97
+ | 0.1175 | 9.26 | 500 | 0.1938 | 0.7045 | 0.6304 | 0.6654 | 0.9640 |
98
+ | 0.1175 | 9.72 | 525 | 0.1880 | 0.7025 | 0.6411 | 0.6704 | 0.9646 |
99
 
100
 
101
  ### Framework versions