<!-- 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. -->
bert-base-uncased-de-ner
This model is a fine-tuned version of bert-base-uncased on the germeval_14 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1374
- Precision: 0.8109
- Recall: 0.7720
- F1: 0.7910
- Accuracy: 0.9786
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
The model was trained on data that follows the IOB convention. Full tagset with indices:
{'O': 0, 'B-PER': 1, 'I-PER': 2, 'B-ORG': 3, 'I-ORG': 4, 'B-LOC': 5, 'I-LOC': 6}
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 0
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.104 | 1.0 | 3000 | 0.0973 | 0.7027 | 0.7323 | 0.7172 | 0.9712 |
| 0.0597 | 2.0 | 6000 | 0.0942 | 0.8135 | 0.7172 | 0.7623 | 0.9766 |
| 0.0345 | 3.0 | 9000 | 0.1051 | 0.7924 | 0.7569 | 0.7742 | 0.9773 |
| 0.0172 | 4.0 | 12000 | 0.1170 | 0.8074 | 0.7628 | 0.7844 | 0.9779 |
| 0.0092 | 5.0 | 15000 | 0.1264 | 0.8068 | 0.7803 | 0.7933 | 0.9788 |
| 0.0035 | 6.0 | 18000 | 0.1374 | 0.8109 | 0.7720 | 0.7910 | 0.9786 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2
