<!-- 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. -->
epochs-xlm-roberta-large-finetuned-conll03-german-2
This model is a fine-tuned version of xlm-roberta-large-finetuned-conll03-german on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4941
- Precision: 0.9294
- Recall: 0.9314
- F1: 0.9304
- Accuracy: 0.9395
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 32
- seed: 852626252
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.9862 | 1.0 | 500 | 0.4789 | 0.8564 | 0.8589 | 0.8576 | 0.8771 |
| 0.4069 | 2.0 | 1000 | 0.3583 | 0.8877 | 0.9016 | 0.8946 | 0.9117 |
| 0.2946 | 3.0 | 1500 | 0.3596 | 0.8947 | 0.8980 | 0.8963 | 0.9142 |
| 0.217 | 4.0 | 2000 | 0.3261 | 0.9189 | 0.9189 | 0.9189 | 0.9302 |
| 0.1666 | 5.0 | 2500 | 0.3089 | 0.9216 | 0.9226 | 0.9221 | 0.9339 |
| 0.1349 | 6.0 | 3000 | 0.3317 | 0.9215 | 0.9235 | 0.9225 | 0.9336 |
| 0.1103 | 7.0 | 3500 | 0.3322 | 0.9191 | 0.9243 | 0.9217 | 0.9340 |
| 0.0898 | 8.0 | 4000 | 0.3764 | 0.9213 | 0.9262 | 0.9237 | 0.9343 |
| 0.0795 | 9.0 | 4500 | 0.3733 | 0.9274 | 0.9290 | 0.9282 | 0.9384 |
| 0.0659 | 10.0 | 5000 | 0.4025 | 0.9269 | 0.9281 | 0.9275 | 0.9368 |
| 0.0579 | 11.0 | 5500 | 0.4470 | 0.9260 | 0.9264 | 0.9262 | 0.9356 |
| 0.0478 | 12.0 | 6000 | 0.4297 | 0.9252 | 0.9283 | 0.9268 | 0.9362 |
| 0.0395 | 13.0 | 6500 | 0.4469 | 0.9284 | 0.9294 | 0.9289 | 0.9386 |
| 0.034 | 14.0 | 7000 | 0.4532 | 0.9269 | 0.9309 | 0.9289 | 0.9381 |
| 0.0302 | 15.0 | 7500 | 0.4678 | 0.9295 | 0.9310 | 0.9303 | 0.9390 |
| 0.0252 | 16.0 | 8000 | 0.4784 | 0.9289 | 0.9312 | 0.9301 | 0.9392 |
| 0.0213 | 17.0 | 8500 | 0.4915 | 0.9288 | 0.9307 | 0.9297 | 0.9388 |
| 0.0202 | 18.0 | 9000 | 0.4936 | 0.9271 | 0.9297 | 0.9284 | 0.9381 |
| 0.0184 | 19.0 | 9500 | 0.4918 | 0.9300 | 0.9307 | 0.9303 | 0.9392 |
| 0.0152 | 20.0 | 10000 | 0.4941 | 0.9294 | 0.9314 | 0.9304 | 0.9395 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
