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: 0.84
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: 1.0312
- Accuracy: 0.84
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: 1
- eval_batch_size: 1
- 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 |
|---|---|---|---|---|
| 1.218 | 1.0 | 899 | 1.4800 | 0.54 |
| 1.2067 | 2.0 | 1798 | 1.6373 | 0.63 |
| 0.0462 | 3.0 | 2697 | 0.9210 | 0.73 |
| 0.0276 | 4.0 | 3596 | 0.9785 | 0.82 |
| 0.0291 | 5.0 | 4495 | 1.2520 | 0.77 |
| 0.0036 | 6.0 | 5394 | 1.1841 | 0.81 |
| 0.0004 | 7.0 | 6293 | 1.1607 | 0.82 |
| 0.0002 | 8.0 | 7192 | 1.2134 | 0.79 |
| 0.0001 | 9.0 | 8091 | 0.9547 | 0.85 |
| 0.0002 | 10.0 | 8990 | 1.0312 | 0.84 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2