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: 1.2425
- 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: 0.0005
- train_batch_size: 8
- eval_batch_size: 12
- 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 |
|---|---|---|---|---|
| 1.2069 | 1.0 | 113 | 1.3361 | 0.5 |
| 1.5776 | 2.0 | 226 | 1.4992 | 0.43 |
| 1.2343 | 3.0 | 339 | 1.2779 | 0.48 |
| 0.8813 | 4.0 | 452 | 1.2418 | 0.62 |
| 0.8836 | 5.0 | 565 | 0.9679 | 0.71 |
| 0.7827 | 6.0 | 678 | 0.9275 | 0.7 |
| 0.4979 | 7.0 | 791 | 1.2511 | 0.69 |
| 0.466 | 8.0 | 904 | 1.0917 | 0.73 |
| 0.6358 | 9.0 | 1017 | 0.7578 | 0.81 |
| 0.5371 | 10.0 | 1130 | 1.2664 | 0.72 |
| 0.0353 | 11.0 | 1243 | 1.2281 | 0.77 |
| 0.0159 | 12.0 | 1356 | 1.5949 | 0.73 |
| 0.0011 | 13.0 | 1469 | 1.0783 | 0.85 |
| 0.0007 | 14.0 | 1582 | 1.3324 | 0.82 |
| 0.0007 | 15.0 | 1695 | 1.2425 | 0.84 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3