metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan-1
results: []
distilhubert-finetuned-gtzan-1
This model is a fine-tuned version of ntu-spml/distilhubert on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.5778
- Accuracy: 0.82
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 14
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.103 | 1.0 | 112 | 2.1288 | 0.42 |
| 1.5948 | 2.0 | 225 | 1.6203 | 0.55 |
| 1.3883 | 3.0 | 337 | 1.2437 | 0.69 |
| 1.1032 | 4.0 | 450 | 1.0490 | 0.73 |
| 0.7595 | 5.0 | 562 | 0.8857 | 0.79 |
| 0.812 | 6.0 | 675 | 0.7776 | 0.8 |
| 0.4903 | 7.0 | 787 | 0.7682 | 0.78 |
| 0.5568 | 8.0 | 900 | 0.7100 | 0.79 |
| 0.405 | 9.0 | 1012 | 0.6279 | 0.84 |
| 0.5888 | 10.0 | 1125 | 0.6944 | 0.8 |
| 0.2576 | 11.0 | 1237 | 0.6027 | 0.83 |
| 0.2123 | 12.0 | 1350 | 0.5891 | 0.83 |
| 0.2008 | 13.0 | 1462 | 0.5659 | 0.83 |
| 0.1343 | 13.94 | 1568 | 0.5778 | 0.82 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3