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: train
split: train
args: train
metrics:
- name: Accuracy
type: accuracy
value: 0.86
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:
- Accuracy: 0.86
- Loss: 0.7644
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: 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: 10
Training results
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|---|---|---|---|---|
| 2.1622 | 1.0 | 113 | 0.36 | 2.0289 |
| 1.6015 | 2.0 | 226 | 0.59 | 1.4290 |
| 1.1929 | 3.0 | 339 | 0.7 | 1.1003 |
| 0.9015 | 4.0 | 452 | 0.76 | 0.8761 |
| 0.7038 | 5.0 | 565 | 0.76 | 0.7516 |
| 0.3261 | 6.0 | 678 | 0.77 | 0.7753 |
| 0.5327 | 7.0 | 791 | 0.79 | 0.6131 |
| 0.1239 | 8.0 | 904 | 0.8 | 0.6283 |
| 0.1193 | 9.0 | 1017 | 0.85 | 0.5770 |
| 0.1405 | 10.0 | 1130 | 0.8 | 0.7979 |
| 0.0113 | 11.0 | 1243 | 0.81 | 0.7830 |
| 0.1392 | 12.0 | 1356 | 0.85 | 0.7350 |
| 0.0065 | 13.0 | 1469 | 0.82 | 0.7935 |
| 0.0049 | 14.0 | 1582 | 0.84 | 0.8323 |
| 0.0041 | 15.0 | 1695 | 0.86 | 0.7644 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1
- Datasets 2.14.0
- Tokenizers 0.13.3