<!-- 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-german-cased-finetuned-7labels
This model is a fine-tuned version of ogimgio/bert-base-german-cased-issues-128 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6844
- Micro f1: 0.6340
- Macro f1: 0.6083
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Micro f1 | Macro f1 |
|---|---|---|---|---|---|
| 0.3574 | 1.0 | 1365 | 0.3370 | 0.6242 | 0.5971 |
| 0.2933 | 2.0 | 2730 | 0.3338 | 0.6385 | 0.6127 |
| 0.2441 | 3.0 | 4095 | 0.3529 | 0.6153 | 0.5938 |
| 0.198 | 4.0 | 5460 | 0.3862 | 0.6255 | 0.6017 |
| 0.158 | 5.0 | 6825 | 0.4100 | 0.6375 | 0.6198 |
| 0.1251 | 6.0 | 8190 | 0.4531 | 0.6431 | 0.6208 |
| 0.0993 | 7.0 | 9555 | 0.5136 | 0.6342 | 0.6117 |
| 0.0784 | 8.0 | 10920 | 0.5632 | 0.6307 | 0.6085 |
| 0.064 | 9.0 | 12285 | 0.6038 | 0.6273 | 0.6010 |
| 0.0535 | 10.0 | 13650 | 0.6519 | 0.6338 | 0.6098 |
| 0.0469 | 11.0 | 15015 | 0.6844 | 0.6340 | 0.6083 |
Framework versions
- Transformers 4.19.0.dev0
- Pytorch 1.11.0+cu102
- Datasets 2.8.0
- Tokenizers 0.12.1
