metadata
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: Deepfake-Audio-Detection-v1
results: []
Deepfake-Audio-Detection-v1
This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0130
- Accuracy: 0.9974
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.0985 | 1.0 | 407 | 0.0883 | 0.9728 |
| 0.0459 | 2.0 | 814 | 0.0351 | 0.9909 |
| 0.004 | 3.0 | 1221 | 0.0389 | 0.9912 |
| 0.0131 | 4.0 | 1628 | 0.0154 | 0.9971 |
| 0.0006 | 5.0 | 2035 | 0.0130 | 0.9974 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0