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: 0.6694
- 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
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- 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 |
---|---|---|---|---|
No log | 0.99 | 56 | 1.9426 | 0.5 |
No log | 1.99 | 112 | 1.4157 | 0.63 |
No log | 2.99 | 168 | 1.1351 | 0.69 |
No log | 3.99 | 224 | 1.0285 | 0.72 |
No log | 4.99 | 280 | 0.8538 | 0.79 |
No log | 5.99 | 336 | 0.8015 | 0.74 |
No log | 6.99 | 392 | 0.6694 | 0.82 |
No log | 7.99 | 448 | 0.6779 | 0.79 |
1.0811 | 8.99 | 504 | 0.6414 | 0.81 |
1.0811 | 9.99 | 560 | 0.6443 | 0.82 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.11.0
- Datasets 2.6.1
- Tokenizers 0.11.6
- Downloads last month
- 19
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.