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: 0.7031
- 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: 16
- eval_batch_size: 16
- 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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.2612 | 1.0 | 57 | 2.2511 | 0.26 |
| 2.1275 | 2.0 | 114 | 2.0384 | 0.36 |
| 1.8071 | 3.0 | 171 | 1.7399 | 0.52 |
| 1.6381 | 4.0 | 228 | 1.5693 | 0.61 |
| 1.4188 | 5.0 | 285 | 1.3573 | 0.61 |
| 1.2974 | 6.0 | 342 | 1.2103 | 0.72 |
| 1.2146 | 7.0 | 399 | 1.1800 | 0.69 |
| 1.0725 | 8.0 | 456 | 1.0126 | 0.77 |
| 1.0492 | 9.0 | 513 | 0.9821 | 0.74 |
| 1.0529 | 10.0 | 570 | 0.9347 | 0.77 |
| 0.895 | 11.0 | 627 | 0.8520 | 0.79 |
| 0.7692 | 12.0 | 684 | 0.8451 | 0.8 |
| 0.6566 | 13.0 | 741 | 0.7763 | 0.82 |
| 0.5885 | 14.0 | 798 | 0.7852 | 0.8 |
| 0.619 | 15.0 | 855 | 0.7443 | 0.8 |
| 0.5572 | 16.0 | 912 | 0.7444 | 0.79 |
| 0.6493 | 17.0 | 969 | 0.7024 | 0.83 |
| 0.5499 | 18.0 | 1026 | 0.7137 | 0.81 |
| 0.5923 | 19.0 | 1083 | 0.7059 | 0.81 |
| 0.5556 | 20.0 | 1140 | 0.7031 | 0.82 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3