<!-- 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. -->
wav2vec2_100k_10s_augmentation_gtzan_model
This model is a fine-tuned version of facebook/wav2vec2-base-100k-voxpopuli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2169
- Accuracy: 0.7438
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: 12
- eval_batch_size: 12
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.2838 | 1.0 | 40 | 2.2550 | 0.35 |
| 2.1032 | 2.0 | 80 | 2.0142 | 0.45 |
| 1.9317 | 3.0 | 120 | 1.8572 | 0.5375 |
| 1.8217 | 4.0 | 160 | 1.7297 | 0.65 |
| 1.6807 | 5.0 | 200 | 1.6411 | 0.6687 |
| 1.6008 | 6.0 | 240 | 1.5568 | 0.6625 |
| 1.4711 | 7.0 | 280 | 1.4869 | 0.7 |
| 1.3914 | 8.0 | 320 | 1.4036 | 0.7188 |
| 1.2889 | 9.0 | 360 | 1.3725 | 0.6813 |
| 1.2285 | 10.0 | 400 | 1.3216 | 0.75 |
| 1.1598 | 11.0 | 440 | 1.2877 | 0.7438 |
| 1.1297 | 12.0 | 480 | 1.2467 | 0.75 |
| 1.0851 | 13.0 | 520 | 1.2242 | 0.7812 |
| 1.0205 | 14.0 | 560 | 1.2413 | 0.7438 |
| 1.0435 | 15.0 | 600 | 1.2169 | 0.7438 |
Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
