|
--- |
|
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: |
|
- Loss: 0.5224 |
|
- Accuracy: 0.88 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 15 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 2.1398 | 1.0 | 56 | 2.1305 | 0.41 | |
|
| 1.6997 | 1.99 | 112 | 1.6665 | 0.55 | |
|
| 1.3111 | 2.99 | 168 | 1.2401 | 0.7 | |
|
| 1.0423 | 4.0 | 225 | 1.1047 | 0.7 | |
|
| 0.8737 | 5.0 | 281 | 0.9386 | 0.71 | |
|
| 0.7303 | 5.99 | 337 | 0.7536 | 0.77 | |
|
| 0.5814 | 6.99 | 393 | 0.7451 | 0.8 | |
|
| 0.4965 | 8.0 | 450 | 0.7118 | 0.76 | |
|
| 0.3755 | 9.0 | 506 | 0.6023 | 0.84 | |
|
| 0.3492 | 9.99 | 562 | 0.6544 | 0.83 | |
|
| 0.1924 | 10.99 | 618 | 0.5607 | 0.88 | |
|
| 0.1424 | 12.0 | 675 | 0.5156 | 0.86 | |
|
| 0.1047 | 13.0 | 731 | 0.5138 | 0.89 | |
|
| 0.0897 | 13.99 | 787 | 0.5660 | 0.84 | |
|
| 0.0819 | 14.93 | 840 | 0.5224 | 0.88 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.30.1 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.4 |
|
- Tokenizers 0.13.3 |
|
|