metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan-v2
results: []
distilhubert-finetuned-gtzan-v2
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: 0.6385
- Accuracy: 0.85
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 13
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.8498 | 1.0 | 113 | 1.8925 | 0.46 |
| 1.2822 | 2.0 | 226 | 1.3237 | 0.65 |
| 1.0384 | 3.0 | 339 | 0.9066 | 0.73 |
| 0.5947 | 4.0 | 452 | 0.6975 | 0.81 |
| 0.4105 | 5.0 | 565 | 0.5710 | 0.87 |
| 0.3001 | 6.0 | 678 | 0.5835 | 0.84 |
| 0.1888 | 7.0 | 791 | 0.5841 | 0.82 |
| 0.2508 | 8.0 | 904 | 0.5339 | 0.84 |
| 0.0914 | 9.0 | 1017 | 0.5488 | 0.86 |
| 0.1202 | 10.0 | 1130 | 0.8281 | 0.83 |
| 0.02 | 11.0 | 1243 | 0.6547 | 0.84 |
| 0.0149 | 12.0 | 1356 | 0.6789 | 0.84 |
| 0.0135 | 13.0 | 1469 | 0.6385 | 0.85 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3