bert-base-german-cased-finetuned-subj_v4_5Epoch
This model is a fine-tuned version of bert-base-german-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3180
- Precision: 0.6640
- Recall: 0.5985
- F1: 0.6296
- Accuracy: 0.8761
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 32 | 0.4471 | 0.4951 | 0.1861 | 0.2706 | 0.8145 |
| No log | 2.0 | 64 | 0.3823 | 0.6096 | 0.4161 | 0.4946 | 0.8472 |
| No log | 3.0 | 96 | 0.3428 | 0.6744 | 0.5292 | 0.5930 | 0.8698 |
| No log | 4.0 | 128 | 0.3216 | 0.6571 | 0.5876 | 0.6204 | 0.8730 |
| No log | 5.0 | 160 | 0.3180 | 0.6640 | 0.5985 | 0.6296 | 0.8761 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.0+cu111
- Datasets 2.0.0
- Tokenizers 0.11.6