music-genre-classifer-20-finetuned-gtzan
This model is a fine-tuned version of facebook/wav2vec2-base on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 1.1602
- Accuracy: 0.81
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: 2e-05
- train_batch_size: 8
- 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: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 2.0297 | 1.0 | 113 | 0.46 | 2.0056 |
| 1.6252 | 2.0 | 226 | 0.61 | 1.5821 |
| 1.4001 | 3.0 | 339 | 0.62 | 1.3967 |
| 1.0201 | 4.0 | 452 | 0.77 | 1.1288 |
| 1.0074 | 5.0 | 565 | 0.69 | 1.0933 |
| 0.8466 | 6.0 | 678 | 0.76 | 0.9162 |
| 0.6966 | 7.0 | 791 | 0.79 | 0.9122 |
| 0.5324 | 8.0 | 904 | 0.82 | 0.7715 |
| 0.6692 | 9.0 | 1017 | 1.0549 | 0.71 |
| 0.7181 | 10.0 | 1130 | 0.8758 | 0.76 |
| 0.5585 | 11.0 | 1243 | 1.0753 | 0.7 |
| 0.4479 | 12.0 | 1356 | 1.1517 | 0.7 |
| 0.3145 | 13.0 | 1469 | 0.8475 | 0.79 |
| 0.8197 | 14.0 | 1582 | 0.8852 | 0.78 |
| 0.4665 | 15.0 | 1695 | 1.0134 | 0.77 |
| 0.2371 | 16.0 | 1808 | 1.0250 | 0.75 |
| 0.3823 | 17.0 | 1921 | 0.9516 | 0.81 |
| 0.5478 | 18.0 | 2034 | 1.2008 | 0.77 |
| 0.3165 | 19.0 | 2147 | 1.0737 | 0.8 |
| 0.1403 | 20.0 | 2260 | 0.9801 | 0.83 |
| 0.2754 | 21.0 | 2373 | 1.0137 | 0.82 |
| 0.2649 | 22.0 | 2486 | 1.2249 | 0.77 |
| 0.0686 | 23.0 | 2599 | 1.3234 | 0.77 |
| 0.0073 | 24.0 | 2712 | 1.2360 | 0.8 |
| 0.0068 | 25.0 | 2825 | 1.1338 | 0.81 |
| 0.0058 | 26.0 | 2938 | 1.2976 | 0.79 |
| 0.0054 | 27.0 | 3051 | 1.1782 | 0.83 |
| 0.0047 | 28.0 | 3164 | 1.0677 | 0.84 |
| 0.0045 | 29.0 | 3277 | 1.1128 | 0.83 |
| 0.0036 | 30.0 | 3390 | 1.1602 | 0.81 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 8
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for vadhri/wav2vec2-base-finetuned-gtzan
Base model
facebook/wav2vec2-base