metadata
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: GTZAN
type: marsyas/gtzan
config: all
split: train
args: all
metrics:
- name: Accuracy
type: accuracy
value: 0.85
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: 0.6405
- 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: 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: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.2068 | 1.0 | 57 | 2.1236 | 0.41 |
| 1.635 | 2.0 | 114 | 1.5471 | 0.57 |
| 1.19 | 3.0 | 171 | 1.1878 | 0.68 |
| 1.0898 | 4.0 | 228 | 1.0190 | 0.71 |
| 0.73 | 5.0 | 285 | 0.8323 | 0.73 |
| 0.6549 | 6.0 | 342 | 0.7693 | 0.76 |
| 0.4567 | 7.0 | 399 | 0.7017 | 0.8 |
| 0.379 | 8.0 | 456 | 0.7082 | 0.79 |
| 0.2807 | 9.0 | 513 | 0.6414 | 0.81 |
| 0.1668 | 10.0 | 570 | 0.6464 | 0.83 |
| 0.167 | 11.0 | 627 | 0.6404 | 0.85 |
| 0.1125 | 12.0 | 684 | 0.6338 | 0.83 |
| 0.0893 | 13.0 | 741 | 0.6447 | 0.86 |
| 0.0604 | 14.0 | 798 | 0.6332 | 0.85 |
| 0.0663 | 15.0 | 855 | 0.6405 | 0.85 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2