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: 0.82
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: 0.6925
- Accuracy: 0.82
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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.9199 | 1.0 | 113 | 1.8344 | 0.49 |
| 1.3377 | 2.0 | 226 | 1.2973 | 0.59 |
| 0.8998 | 3.0 | 339 | 0.9938 | 0.67 |
| 0.6958 | 4.0 | 452 | 0.8983 | 0.69 |
| 0.4815 | 5.0 | 565 | 0.7645 | 0.73 |
| 0.2497 | 6.0 | 678 | 0.7328 | 0.76 |
| 0.2588 | 7.0 | 791 | 0.8284 | 0.73 |
| 0.1133 | 8.0 | 904 | 0.7011 | 0.78 |
| 0.1659 | 9.0 | 1017 | 0.6872 | 0.78 |
| 0.0656 | 10.0 | 1130 | 0.6925 | 0.82 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2