metadata
library_name: transformers
license: apache-2.0
base_model: PKOBP/polish-roberta-8k
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: mwik-classifier-ext
results: []
mwik-classifier-ext
This model is a fine-tuned version of PKOBP/polish-roberta-8k on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1140
- Accuracy: 0.7371
- Precision: 0.7338
- Recall: 0.7371
- F1: 0.7276
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: 0.0001
- train_batch_size: 24
- eval_batch_size: 48
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 96
- 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: polynomial
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 3.0764 | 1.0 | 64 | 1.6987 | 0.6210 | 0.5278 | 0.6210 | 0.5529 |
| 1.6258 | 2.0 | 128 | 1.3065 | 0.6845 | 0.6331 | 0.6845 | 0.6401 |
| 1.2496 | 3.0 | 192 | 1.1332 | 0.7113 | 0.6755 | 0.7113 | 0.6801 |
| 0.7597 | 4.0 | 256 | 1.0614 | 0.7435 | 0.7298 | 0.7435 | 0.7219 |
| 0.6033 | 5.0 | 320 | 1.0464 | 0.7565 | 0.7427 | 0.7565 | 0.7423 |
| 0.457 | 6.0 | 384 | 1.0559 | 0.7496 | 0.7399 | 0.7496 | 0.7383 |
| 0.3654 | 7.0 | 448 | 1.0642 | 0.7519 | 0.7393 | 0.7519 | 0.7395 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1