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: 0.81
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: 0.6504
- Accuracy: 0.81
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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|
| 2.2428 | 1.0 | 113 | 2.1981 | 0.35 |
| 1.7527 | 2.0 | 226 | 1.7611 | 0.55 |
| 1.6002 | 3.0 | 339 | 1.4516 | 0.65 |
| 1.2101 | 4.0 | 452 | 1.2245 | 0.7 |
| 1.1006 | 5.0 | 565 | 1.0758 | 0.73 |
| 0.9583 | 6.0 | 678 | 0.9477 | 0.76 |
| 0.8705 | 7.0 | 791 | 0.8907 | 0.77 |
| 0.6892 | 8.0 | 904 | 0.8438 | 0.75 |
| 0.6141 | 9.0 | 1017 | 0.7574 | 0.79 |
| 0.5614 | 10.0 | 1130 | 0.7300 | 0.81 |
| 0.5347 | 11.0 | 1243 | 0.6830 | 0.8 |
| 0.5106 | 12.0 | 1356 | 0.7286 | 0.81 |
| 0.4662 | 13.0 | 1469 | 0.6701 | 0.8 |
| 0.6223 | 14.0 | 1582 | 0.6728 | 0.8 |
| 0.3604 | 15.0 | 1695 | 0.6504 | 0.81 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1