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: 0.6925
- Accuracy: 0.83
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: 0.0001115511981046745
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- 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: 9
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 1.278 | 1.0 | 112 | 0.57 | 1.3298 |
| 0.8315 | 2.0 | 225 | 0.73 | 0.9432 |
| 0.7709 | 3.0 | 337 | 0.72 | 0.9310 |
| 0.5427 | 4.0 | 450 | 0.72 | 0.8738 |
| 0.2645 | 4.98 | 560 | 0.79 | 0.6648 |
| 0.245 | 6.0 | 672 | 0.83 | 0.6147 |
| 0.1331 | 6.99 | 784 | 0.83 | 0.6305 |
| 0.1863 | 8.0 | 896 | 0.6356 | 0.84 |
| 0.0843 | 8.99 | 1008 | 0.6925 | 0.83 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3