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: default
split: train
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.87
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: 0.5133
- Accuracy: 0.87
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: 12
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.0567 | 1.0 | 113 | 1.8854 | 0.53 |
| 1.3125 | 2.0 | 226 | 1.2433 | 0.65 |
| 1.0641 | 3.0 | 339 | 0.8965 | 0.79 |
| 0.8328 | 4.0 | 452 | 0.8620 | 0.71 |
| 0.6089 | 5.0 | 565 | 0.6442 | 0.82 |
| 0.3886 | 6.0 | 678 | 0.5787 | 0.82 |
| 0.4292 | 7.0 | 791 | 0.5542 | 0.84 |
| 0.1236 | 8.0 | 904 | 0.4888 | 0.86 |
| 0.1899 | 9.0 | 1017 | 0.4964 | 0.85 |
| 0.0825 | 10.0 | 1130 | 0.4998 | 0.9 |
| 0.0488 | 11.0 | 1243 | 0.5106 | 0.88 |
| 0.0353 | 12.0 | 1356 | 0.5133 | 0.87 |
Framework versions
- Transformers 4.32.0
- Pytorch 1.11.0+cu113
- Datasets 2.14.4
- Tokenizers 0.13.3