wav2vec2-object
This model is a fine-tuned version of facebook/wav2vec2-base on the MatsRooth/object dataset. It achieves the following results on the evaluation set:
- Loss: 0.1285
- Accuracy: 0.9836
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: 32
- seed: 0
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.6367 | 1.0 | 84 | 1.2848 | 0.9335 |
| 0.7071 | 2.0 | 168 | 0.5108 | 0.9683 |
| 0.4246 | 3.0 | 252 | 0.3018 | 0.9747 |
| 0.3173 | 4.0 | 336 | 0.2098 | 0.9768 |
| 0.2585 | 5.0 | 420 | 0.1685 | 0.9784 |
| 0.192 | 6.0 | 504 | 0.1285 | 0.9836 |
| 0.1772 | 7.0 | 588 | 0.1210 | 0.9810 |
| 0.1435 | 8.0 | 672 | 0.1141 | 0.9805 |
| 0.1425 | 9.0 | 756 | 0.1094 | 0.9836 |
| 0.1398 | 10.0 | 840 | 0.1077 | 0.9826 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.13.1
- Tokenizers 0.15.0