| --- |
| library_name: transformers |
| base_model: Kkonjeong/wav2vec2-base-korean |
| tags: |
| - generated_from_trainer |
| metrics: |
| - accuracy |
| model-index: |
| - name: audio_cls_ko |
| 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. --> |
|
|
| # audio_cls_ko |
|
|
| This model is a fine-tuned version of [Kkonjeong/wav2vec2-base-korean](https://huggingface.co/Kkonjeong/wav2vec2-base-korean) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.5750 |
| - Accuracy: 0.8655 |
|
|
| ## 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: 8 |
| - eval_batch_size: 8 |
| - seed: 42 |
| - gradient_accumulation_steps: 2 |
| - total_train_batch_size: 16 |
| - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
| - lr_scheduler_type: linear |
| - lr_scheduler_warmup_ratio: 0.1 |
| - num_epochs: 10 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Accuracy | |
| |:-------------:|:-----:|:----:|:---------------:|:--------:| |
| | 2.65 | 1.0 | 30 | 2.6718 | 0.0420 | |
| | 2.3368 | 2.0 | 60 | 2.1884 | 0.4706 | |
| | 1.8532 | 3.0 | 90 | 1.7631 | 0.4790 | |
| | 1.3174 | 4.0 | 120 | 1.3977 | 0.6723 | |
| | 0.941 | 5.0 | 150 | 1.1978 | 0.7731 | |
| | 0.5739 | 6.0 | 180 | 0.9341 | 0.7815 | |
| | 0.4066 | 7.0 | 210 | 0.8123 | 0.8151 | |
| | 0.2582 | 8.0 | 240 | 0.6095 | 0.8992 | |
| | 0.1904 | 9.0 | 270 | 0.5689 | 0.8824 | |
| | 0.1638 | 10.0 | 300 | 0.5750 | 0.8655 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.56.1 |
| - Pytorch 2.8.0+cu126 |
| - Datasets 4.0.0 |
| - Tokenizers 0.22.0 |
| |
