license: apache-2.0 | |
base_model: google/vit-base-patch16-224 | |
model-index: | |
- name: vit-base-vocalsound | |
results: [] | |
# vit-base-vocalsound | |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on [VocalSound](https://github.com/YuanGongND/vocalsound) dataset. | |
It achieves the following results on the evaluation set: | |
- accuracy: 81.5 | |
- precision (micro): 86.9 | |
- recall (micro): 76.4 | |
- f1 score (micro): 81.3 | |
- f1 score (macro): 81.2 | |
## Training and evaluation data | |
Training: VocalSound training split (#samples = 15570) | |
Evaluation: VocalSound test split(#samples = 3594) | |
### Training hyperparameters | |
The following hyperparameters were used during training: | |
- optimizer: AdamW | |
- weight_decay: 0 | |
- learning_rate: 5e-5 | |
- batch_size: 32 | |
- training_precision: float32 | |
### Framework versions | |
- Transformers 4.27.4 | |
- TensorFlow 2.12.0 | |
- Tokenizers 0.13.3 |