distilhubert-finetuned-gtzan-v3
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: 0.5752
- Accuracy: 0.83
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: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.9108 | 1.0 | 113 | 1.9472 | 0.43 |
1.3286 | 2.0 | 226 | 1.4173 | 0.65 |
1.032 | 3.0 | 339 | 0.9815 | 0.67 |
0.726 | 4.0 | 452 | 0.7403 | 0.79 |
0.4621 | 5.0 | 565 | 0.6390 | 0.8 |
0.3439 | 6.0 | 678 | 0.5248 | 0.85 |
0.1592 | 7.0 | 791 | 0.4861 | 0.86 |
0.1283 | 8.0 | 904 | 0.4995 | 0.87 |
0.1191 | 9.0 | 1017 | 0.4804 | 0.87 |
0.0236 | 10.0 | 1130 | 0.6737 | 0.8 |
0.0146 | 11.0 | 1243 | 0.6211 | 0.81 |
0.0105 | 12.0 | 1356 | 0.5806 | 0.86 |
0.008 | 13.0 | 1469 | 0.5645 | 0.84 |
0.0082 | 14.0 | 1582 | 0.6033 | 0.83 |
0.0072 | 15.0 | 1695 | 0.5752 | 0.83 |
Framework versions
- Transformers 4.30.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 10
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.