distilhubert-finetuned-mgs
This model is a fine-tuned version of ntu-spml/distilhubert on the music_genres_small dataset. It achieves the following results on the evaluation set:
- Loss: 1.6640
- Accuracy: 0.43
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.0210 | 1.0 | 113 | 2.0817 | 0.28 |
| 1.8568 | 2.0 | 226 | 1.8664 | 0.29 |
| 1.9320 | 3.0 | 339 | 1.7749 | 0.36 |
| 1.4782 | 4.0 | 452 | 1.7151 | 0.29 |
| 1.4109 | 5.0 | 565 | 1.6726 | 0.36 |
| 1.2085 | 6.0 | 678 | 1.5865 | 0.39 |
| 1.2247 | 7.0 | 791 | 1.6049 | 0.41 |
| 1.0329 | 8.0 | 904 | 1.7045 | 0.39 |
| 0.7782 | 9.0 | 1017 | 1.6659 | 0.44 |
| 0.8732 | 10.0 | 1130 | 1.6640 | 0.43 |
Framework versions
- Transformers 5.10.2
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2