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
metrics:
- name: Accuracy
type: accuracy
value: 0.8
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: 0.7861
- Accuracy: 0.8
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 |
|---|---|---|---|---|
| 1.9073 | 1.0 | 113 | 1.8699 | 0.4 |
| 1.3144 | 2.0 | 226 | 1.2309 | 0.625 |
| 0.8747 | 3.0 | 339 | 0.9318 | 0.74 |
| 0.6776 | 4.0 | 452 | 0.8368 | 0.735 |
| 0.383 | 5.0 | 565 | 0.6930 | 0.745 |
| 0.3383 | 6.0 | 678 | 0.8012 | 0.755 |
| 0.2922 | 7.0 | 791 | 0.6724 | 0.78 |
| 0.1086 | 8.0 | 904 | 0.7984 | 0.755 |
| 0.0409 | 9.0 | 1017 | 0.7385 | 0.805 |
| 0.0507 | 10.0 | 1130 | 0.6669 | 0.805 |
| 0.0424 | 11.0 | 1243 | 0.7698 | 0.815 |
| 0.0078 | 12.0 | 1356 | 0.7985 | 0.81 |
| 0.0068 | 13.0 | 1469 | 0.7679 | 0.81 |
| 0.0063 | 14.0 | 1582 | 0.8139 | 0.795 |
| 0.0065 | 15.0 | 1695 | 0.7861 | 0.8 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.15.2