|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- marsyas/gtzan |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: distilhubert-finetuned-gtzan |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# distilhubert-finetuned-gtzan |
|
|
|
This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the GTZAN dataset. |
|
It achieves the following results on the evaluation set: |
|
- Accuracy: 0.87 |
|
- Loss: 0.6300 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 3 |
|
- total_train_batch_size: 6 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 10 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Accuracy | Validation Loss | |
|
|:-------------:|:-----:|:----:|:--------:|:---------------:| |
|
| 1.8192 | 1.0 | 150 | 0.41 | 1.7706 | |
|
| 1.1745 | 2.0 | 300 | 0.64 | 1.2586 | |
|
| 0.7885 | 3.0 | 450 | 0.76 | 0.8419 | |
|
| 0.661 | 4.0 | 600 | 0.81 | 0.7019 | |
|
| 0.3334 | 5.0 | 750 | 0.84 | 0.5766 | |
|
| 0.3265 | 6.0 | 900 | 0.83 | 0.5862 | |
|
| 0.0928 | 7.0 | 1050 | 0.88 | 0.5613 | |
|
| 0.0698 | 8.0 | 1200 | 0.86 | 0.6432 | |
|
| 0.0562 | 9.0 | 1350 | 0.87 | 0.6141 | |
|
| 0.0268 | 10.0 | 1500 | 0.87 | 0.6300 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.29.2 |
|
- Pytorch 2.0.1 |
|
- Datasets 2.13.1 |
|
- Tokenizers 0.13.3 |
|
|
