encoder / README.md
swtb's picture
End of training
bd34ad8 verified
---
license: mit
base_model: xlm-roberta-large
tags:
- generated_from_trainer
datasets:
- conll2003
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: encoder
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: 0.922421290659245
- name: Recall
type: recall
value: 0.9389164305949008
- name: F1
type: f1
value: 0.9305957708168815
- name: Accuracy
type: accuracy
value: 0.9842790998169484
---
<!-- 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. -->
# encoder
This model is a fine-tuned version of [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) on the conll2003 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2344
- Precision: 0.9224
- Recall: 0.9389
- F1: 0.9306
- Accuracy: 0.9843
## 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 |
|:-------------:|:------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
| 0.657 | 0.3333 | 1441 | 0.1806 | 0.8261 | 0.8383 | 0.8322 | 0.9700 |
| 0.0884 | 0.6667 | 2882 | 0.1383 | 0.8822 | 0.8913 | 0.8867 | 0.9783 |
| 0.0637 | 1.0 | 4323 | 0.1343 | 0.9032 | 0.9132 | 0.9082 | 0.9811 |
| 0.0427 | 1.3333 | 5764 | 0.1527 | 0.9014 | 0.9210 | 0.9111 | 0.9817 |
| 0.0412 | 1.6667 | 7205 | 0.1450 | 0.9109 | 0.9301 | 0.9204 | 0.9838 |
| 0.0391 | 2.0 | 8646 | 0.1459 | 0.9145 | 0.9221 | 0.9183 | 0.9832 |
| 0.0235 | 2.3333 | 10087 | 0.1848 | 0.9041 | 0.9313 | 0.9175 | 0.9821 |
| 0.0228 | 2.6667 | 11528 | 0.1539 | 0.9188 | 0.9375 | 0.9281 | 0.9846 |
| 0.0283 | 3.0 | 12969 | 0.1513 | 0.9137 | 0.9295 | 0.9215 | 0.9833 |
| 0.0176 | 3.3333 | 14410 | 0.1748 | 0.9232 | 0.9347 | 0.9289 | 0.9842 |
| 0.0177 | 3.6667 | 15851 | 0.1706 | 0.9234 | 0.9331 | 0.9282 | 0.9848 |
| 0.0191 | 4.0 | 17292 | 0.1784 | 0.9095 | 0.9309 | 0.9201 | 0.9829 |
| 0.0131 | 4.3333 | 18733 | 0.1862 | 0.9130 | 0.9361 | 0.9244 | 0.9833 |
| 0.0138 | 4.6667 | 20174 | 0.1883 | 0.9133 | 0.9322 | 0.9226 | 0.9827 |
| 0.0128 | 5.0 | 21615 | 0.1986 | 0.9104 | 0.9304 | 0.9203 | 0.9820 |
| 0.0112 | 5.3333 | 23056 | 0.2002 | 0.9172 | 0.9356 | 0.9263 | 0.9833 |
| 0.0097 | 5.6667 | 24497 | 0.1784 | 0.9257 | 0.9394 | 0.9325 | 0.9846 |
| 0.0068 | 6.0 | 25938 | 0.1929 | 0.9210 | 0.9333 | 0.9271 | 0.9838 |
| 0.0068 | 6.3333 | 27379 | 0.2086 | 0.9212 | 0.9382 | 0.9296 | 0.9840 |
| 0.0057 | 6.6667 | 28820 | 0.2035 | 0.9240 | 0.9368 | 0.9304 | 0.9844 |
| 0.006 | 7.0 | 30261 | 0.2098 | 0.9198 | 0.9379 | 0.9287 | 0.9841 |
| 0.0042 | 7.3333 | 31702 | 0.2236 | 0.9182 | 0.9327 | 0.9254 | 0.9835 |
| 0.0054 | 7.6667 | 33143 | 0.2267 | 0.9196 | 0.9361 | 0.9278 | 0.9833 |
| 0.0029 | 8.0 | 34584 | 0.2162 | 0.9257 | 0.9375 | 0.9316 | 0.9846 |
| 0.0022 | 8.3333 | 36025 | 0.2120 | 0.9241 | 0.9403 | 0.9322 | 0.9849 |
| 0.0045 | 8.6667 | 37466 | 0.2185 | 0.9247 | 0.9393 | 0.9319 | 0.9846 |
| 0.0029 | 9.0 | 38907 | 0.2182 | 0.9247 | 0.9387 | 0.9316 | 0.9846 |
| 0.0021 | 9.3333 | 40348 | 0.2316 | 0.9231 | 0.9394 | 0.9312 | 0.9842 |
| 0.002 | 9.6667 | 41789 | 0.2358 | 0.9226 | 0.9387 | 0.9306 | 0.9842 |
| 0.0019 | 10.0 | 43230 | 0.2344 | 0.9224 | 0.9389 | 0.9306 | 0.9843 |
### Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1