metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan
results: []
distilhubert-finetuned-gtzan
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: 0.6711
- 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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1962 | 1.0 | 113 | 2.2220 | 0.29 |
| 1.9431 | 2.0 | 226 | 1.8877 | 0.5 |
| 1.634 | 3.0 | 339 | 1.5106 | 0.63 |
| 1.3403 | 4.0 | 452 | 1.3191 | 0.66 |
| 1.1067 | 5.0 | 565 | 1.1082 | 0.68 |
| 1.0416 | 6.0 | 678 | 1.0664 | 0.72 |
| 0.7723 | 7.0 | 791 | 0.9729 | 0.77 |
| 0.8281 | 8.0 | 904 | 0.8799 | 0.78 |
| 0.6344 | 9.0 | 1017 | 0.8142 | 0.77 |
| 0.8819 | 10.0 | 1130 | 0.8719 | 0.73 |
| 0.4279 | 11.0 | 1243 | 0.8150 | 0.78 |
| 0.425 | 12.0 | 1356 | 0.7137 | 0.81 |
| 0.2749 | 13.0 | 1469 | 0.6987 | 0.8 |
| 0.2182 | 14.0 | 1582 | 0.6849 | 0.82 |
| 0.2128 | 15.0 | 1695 | 0.6918 | 0.82 |
| 0.1831 | 16.0 | 1808 | 0.6600 | 0.81 |
| 0.1517 | 17.0 | 1921 | 0.6571 | 0.82 |
| 0.2888 | 18.0 | 2034 | 0.6880 | 0.81 |
| 0.1605 | 19.0 | 2147 | 0.6874 | 0.82 |
| 0.1492 | 20.0 | 2260 | 0.6711 | 0.82 |
Framework versions
- Transformers 4.29.2
- Pytorch 1.13.1+cu117
- Datasets 2.12.0
- Tokenizers 0.13.3