metadata
license: apache-2.0
base_model: ntu-spml/distilhubert
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan2
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: 0.7125
distilhubert-finetuned-gtzan2
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: 1.5220
- Accuracy: 0.7125
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: 0.001
- train_batch_size: 32
- eval_batch_size: 32
- 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: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.7489 | 1.0 | 29 | 1.4959 | 0.3875 |
| 1.328 | 2.0 | 58 | 2.0243 | 0.35 |
| 1.2168 | 3.0 | 87 | 1.1332 | 0.5875 |
| 1.0299 | 4.0 | 116 | 1.4826 | 0.5375 |
| 0.911 | 5.0 | 145 | 1.2510 | 0.625 |
| 1.0819 | 6.0 | 174 | 1.7365 | 0.55 |
| 0.9513 | 7.0 | 203 | 1.3000 | 0.6 |
| 0.5687 | 8.0 | 232 | 1.0503 | 0.7125 |
| 0.4684 | 9.0 | 261 | 1.1167 | 0.7125 |
| 0.2836 | 10.0 | 290 | 1.5990 | 0.65 |
| 0.138 | 11.0 | 319 | 1.2096 | 0.7375 |
| 0.0406 | 12.0 | 348 | 1.7311 | 0.6375 |
| 0.0341 | 13.0 | 377 | 1.7048 | 0.6375 |
| 0.0059 | 14.0 | 406 | 1.4933 | 0.7 |
| 0.0034 | 15.0 | 435 | 1.5220 | 0.7125 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2