metadata
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: 25-distilhubert-finetuned-gtzan
results:
- task:
name: Audio Classification
type: audio-classification
dataset:
name: GTZAN
type: marsyas/gtzan
config: default
split: train[:25%]
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.96
25-distilhubert-finetuned-gtzan
This model is a small fine-tuned version of ntu-spml/distilhubert on the GTZAN dataset. It takes 25% of the GTZAN dataset to fine-tune. It achieves the following results on the evaluation set:
- Loss: 0.1955
- Accuracy: 0.96
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
- 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 | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3123 | 1.0 | 57 | 1.3463 | 0.56 |
| 0.6113 | 2.0 | 114 | 0.5379 | 0.96 |
| 0.1295 | 3.0 | 171 | 0.2719 | 0.96 |
| 0.2878 | 4.0 | 228 | 0.0979 | 0.96 |
| 0.0168 | 5.0 | 285 | 0.1527 | 0.96 |
| 0.0104 | 6.0 | 342 | 0.2320 | 0.96 |
| 0.0067 | 7.0 | 399 | 0.1798 | 0.96 |
| 0.0051 | 8.0 | 456 | 0.1827 | 0.96 |
| 0.0041 | 9.0 | 513 | 0.1918 | 0.96 |
| 0.0055 | 10.0 | 570 | 0.1955 | 0.96 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2