superb_ks_42
This model is a fine-tuned version of facebook/wav2vec2-base on the superb dataset. It achieves the following results on the evaluation set:
- Loss: 77.6746
- Accuracy: 0.6233
- Test Accuracy: 0.6233
- Df Accuracy: 0.1373
- Unlearn Overall Accuracy: 0.7430
- Unlearn Time: 756.9285
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: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2.5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Overall Accuracy | Unlearn Overall Accuracy | Time |
|---|---|---|---|---|---|---|---|
| 0.0 | 1.0 | 1440 | 77.6746 | 0.1373 | 0.7430 | 0.7430 | -1 |
| 0.0 | 2.0 | 2880 | 168.1211 | 0.1359 | 0.7426 | 0.7426 | -1 |
| 0.0 | 2.5 | 3600 | 171.9577 | 0.1357 | 0.7424 | 0.7424 | -1 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.2