metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- marsyas/gtzan
metrics:
- accuracy
model-index:
- name: distilhubert-finetuned-gtzan
results: []
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: 0.5382
- Accuracy: 0.88
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.7451 | 1.0 | 113 | 1.7870 | 0.46 |
| 1.3112 | 2.0 | 226 | 1.3802 | 0.6 |
| 1.0303 | 3.0 | 339 | 1.0234 | 0.71 |
| 0.7231 | 4.0 | 452 | 0.8570 | 0.75 |
| 0.5857 | 5.0 | 565 | 0.7428 | 0.78 |
| 0.3953 | 6.0 | 678 | 0.6162 | 0.85 |
| 0.2594 | 7.0 | 791 | 0.6163 | 0.84 |
| 0.3199 | 8.0 | 904 | 0.5505 | 0.85 |
| 0.1998 | 9.0 | 1017 | 0.5385 | 0.86 |
| 0.2329 | 10.0 | 1130 | 0.5382 | 0.88 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3