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: 0.6477
- 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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 12
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.1618 | 1.0 | 75 | 2.0497 | 0.36 |
| 1.5327 | 2.0 | 150 | 1.4568 | 0.62 |
| 1.1622 | 3.0 | 225 | 1.1626 | 0.66 |
| 0.849 | 4.0 | 300 | 0.9894 | 0.74 |
| 0.6072 | 5.0 | 375 | 0.8128 | 0.75 |
| 0.4014 | 6.0 | 450 | 0.7118 | 0.79 |
| 0.3285 | 7.0 | 525 | 0.7482 | 0.83 |
| 0.3074 | 8.0 | 600 | 0.5633 | 0.85 |
| 0.242 | 9.0 | 675 | 0.6613 | 0.82 |
| 0.069 | 10.0 | 750 | 0.5173 | 0.85 |
| 0.1281 | 11.0 | 825 | 0.6102 | 0.83 |
| 0.0334 | 12.0 | 900 | 0.5990 | 0.84 |
| 0.0307 | 13.0 | 975 | 0.6227 | 0.86 |
| 0.0339 | 14.0 | 1050 | 0.6331 | 0.85 |
| 0.0239 | 15.0 | 1125 | 0.6477 | 0.85 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.2
- Tokenizers 0.13.3