|
--- |
|
library_name: transformers |
|
license: cc-by-4.0 |
|
base_model: vesteinn/DanskBERT |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
- f1 |
|
model-index: |
|
- name: danskbert_indirect_speech |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# danskbert_indirect_speech |
|
|
|
This model is a fine-tuned version of [vesteinn/DanskBERT](https://huggingface.co/vesteinn/DanskBERT) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Accuracy: 0.6698 |
|
- Precision: 0.7042 |
|
- Recall: 0.6698 |
|
- F1: 0.6640 |
|
- Loss: 0.7715 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 8 |
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- num_epochs: 20 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Accuracy | Precision | Recall | F1 | Validation Loss | |
|
|:-------------:|:-------:|:----:|:--------:|:---------:|:------:|:------:|:---------------:| |
|
| No log | 1.0 | 9 | 0.3820 | 0.1459 | 0.3820 | 0.2112 | 1.1479 | |
|
| No log | 2.0 | 18 | 0.3820 | 0.1459 | 0.3820 | 0.2112 | 1.4862 | |
|
| No log | 3.0 | 27 | 0.5385 | 0.2900 | 0.5385 | 0.3770 | 1.0168 | |
|
| No log | 4.0 | 36 | 0.3820 | 0.1459 | 0.3820 | 0.2112 | 1.0134 | |
|
| No log | 5.0 | 45 | 0.4551 | 0.6492 | 0.4551 | 0.3461 | 0.9242 | |
|
| No log | 6.0 | 54 | 0.6217 | 0.6269 | 0.6217 | 0.5716 | 0.7808 | |
|
| No log | 7.0 | 63 | 0.6183 | 0.6601 | 0.6183 | 0.5446 | 0.7970 | |
|
| No log | 8.0 | 72 | 0.4519 | 0.6933 | 0.4519 | 0.3345 | 1.0565 | |
|
| No log | 9.0 | 81 | 0.7000 | 0.6985 | 0.7000 | 0.6842 | 0.7125 | |
|
| No log | 10.0 | 90 | 0.6480 | 0.6978 | 0.6480 | 0.6395 | 0.7874 | |
|
| No log | 11.0 | 99 | 0.6226 | 0.7064 | 0.6226 | 0.6062 | 0.8571 | |
|
| No log | 12.0 | 108 | 0.5364 | 0.7104 | 0.5364 | 0.4812 | 1.1975 | |
|
| No log | 13.0 | 117 | 0.7423 | 0.7327 | 0.7423 | 0.7336 | 0.6509 | |
|
| No log | 14.0 | 126 | 0.7372 | 0.7275 | 0.7372 | 0.7306 | 0.6489 | |
|
| No log | 15.0 | 135 | 0.5954 | 0.7039 | 0.5954 | 0.5712 | 0.9821 | |
|
| No log | 16.0 | 144 | 0.7372 | 0.7317 | 0.7372 | 0.7258 | 0.6768 | |
|
| No log | 17.0 | 153 | 0.6457 | 0.7049 | 0.6457 | 0.6355 | 0.8276 | |
|
| No log | 17.8235 | 160 | 0.6698 | 0.7042 | 0.6698 | 0.6640 | 0.7715 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.47.1 |
|
- Pytorch 2.5.1+cu124 |
|
- Tokenizers 0.21.0 |
|
|
