metadata
license: mit
base_model: dbmdz/bert-base-german-uncased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: class_classificator_results
results: []
class_classificator_results
This model is a fine-tuned version of dbmdz/bert-base-german-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7400
- Precision: 0.9096
- Recall: 0.9096
- F1: 0.9096
- Accuracy: 0.9096
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: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 1.321 | 1.0 | 2527 | 1.1812 | 0.8505 | 0.8505 | 0.8505 | 0.8505 |
| 0.8803 | 2.0 | 5054 | 0.8872 | 0.8850 | 0.8850 | 0.8850 | 0.8850 |
| 0.6046 | 3.0 | 7581 | 0.7477 | 0.9042 | 0.9042 | 0.9042 | 0.9042 |
| 0.4113 | 4.0 | 10108 | 0.7400 | 0.9096 | 0.9096 | 0.9096 | 0.9096 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.0.1+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1