File size: 4,853 Bytes
384ee51
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
---
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: XLM-RoBERTa-Base-Conll2003-English-NER-Finetune
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: conll2003
      type: conll2003
      config: conll2003
      split: test
      args: conll2003
    metrics:
    - name: Precision
      type: precision
      value: 0.9051348999129678
    - name: Recall
      type: recall
      value: 0.9206798866855525
    - name: F1
      type: f1
      value: 0.9128412182919337
    - name: Accuracy
      type: accuracy
      value: 0.9819532680090449
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# XLM-RoBERTa-Base-Conll2003-English-NER-Finetune

This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the conll2003 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1887
- Precision: 0.9051
- Recall: 0.9207
- F1: 0.9128
- Accuracy: 0.9820

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 10

### Training results

| Training Loss | Epoch  | Step  | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.6956        | 0.3333 | 1441  | 0.1847          | 0.5575    | 0.5845 | 0.5707 | 0.9322   |
| 0.1126        | 0.6667 | 2882  | 0.1192          | 0.8533    | 0.8762 | 0.8646 | 0.9748   |
| 0.0678        | 1.0    | 4323  | 0.1404          | 0.8551    | 0.8899 | 0.8721 | 0.9756   |
| 0.0528        | 1.3333 | 5764  | 0.1332          | 0.8868    | 0.9040 | 0.8953 | 0.9800   |
| 0.0523        | 1.6667 | 7205  | 0.1352          | 0.8868    | 0.9083 | 0.8974 | 0.9800   |
| 0.0494        | 2.0    | 8646  | 0.1437          | 0.8855    | 0.9063 | 0.8958 | 0.9793   |
| 0.0351        | 2.3333 | 10087 | 0.1592          | 0.8867    | 0.9092 | 0.8978 | 0.9794   |
| 0.0341        | 2.6667 | 11528 | 0.1532          | 0.8919    | 0.9131 | 0.9024 | 0.9801   |
| 0.034         | 3.0    | 12969 | 0.1404          | 0.8967    | 0.9155 | 0.9060 | 0.9808   |
| 0.024         | 3.3333 | 14410 | 0.1601          | 0.8978    | 0.9145 | 0.9061 | 0.9805   |
| 0.0267        | 3.6667 | 15851 | 0.1563          | 0.9047    | 0.9180 | 0.9113 | 0.9819   |
| 0.0255        | 4.0    | 17292 | 0.1406          | 0.9093    | 0.9193 | 0.9142 | 0.9827   |
| 0.0199        | 4.3333 | 18733 | 0.1604          | 0.9047    | 0.9225 | 0.9135 | 0.9821   |
| 0.0187        | 4.6667 | 20174 | 0.1541          | 0.9106    | 0.9251 | 0.9178 | 0.9829   |
| 0.0169        | 5.0    | 21615 | 0.1692          | 0.9009    | 0.9163 | 0.9085 | 0.9814   |
| 0.0159        | 5.3333 | 23056 | 0.1738          | 0.9012    | 0.9205 | 0.9107 | 0.9817   |
| 0.0141        | 5.6667 | 24497 | 0.1610          | 0.9039    | 0.9178 | 0.9108 | 0.9821   |
| 0.0141        | 6.0    | 25938 | 0.1797          | 0.8977    | 0.9164 | 0.9070 | 0.9805   |
| 0.0105        | 6.3333 | 27379 | 0.1707          | 0.9026    | 0.9187 | 0.9106 | 0.9821   |
| 0.0104        | 6.6667 | 28820 | 0.1832          | 0.9036    | 0.9191 | 0.9113 | 0.9812   |
| 0.0135        | 7.0    | 30261 | 0.1743          | 0.9024    | 0.9214 | 0.9118 | 0.9817   |
| 0.0101        | 7.3333 | 31702 | 0.1877          | 0.9006    | 0.9194 | 0.9099 | 0.9812   |
| 0.0113        | 7.6667 | 33143 | 0.1893          | 0.9009    | 0.9187 | 0.9097 | 0.9811   |
| 0.0088        | 8.0    | 34584 | 0.1867          | 0.9050    | 0.9196 | 0.9123 | 0.9818   |
| 0.0068        | 8.3333 | 36025 | 0.1901          | 0.9022    | 0.9182 | 0.9101 | 0.9812   |
| 0.0088        | 8.6667 | 37466 | 0.1956          | 0.9037    | 0.9193 | 0.9114 | 0.9813   |
| 0.0085        | 9.0    | 38907 | 0.1873          | 0.9055    | 0.9216 | 0.9135 | 0.9820   |
| 0.0068        | 9.3333 | 40348 | 0.1922          | 0.9049    | 0.9217 | 0.9133 | 0.9817   |
| 0.006         | 9.6667 | 41789 | 0.1915          | 0.9047    | 0.9214 | 0.9130 | 0.9817   |
| 0.006         | 10.0   | 43230 | 0.1887          | 0.9051    | 0.9207 | 0.9128 | 0.9820   |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1