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: 0.7186
- Accuracy: 0.86
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: 0.0001
- train_batch_size: 8
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-05
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 12
- label_smoothing_factor: 0.05
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.7045 | 1.0 | 113 | 1.7952 | 0.44 |
1.1808 | 2.0 | 226 | 1.1510 | 0.66 |
1.0978 | 3.0 | 339 | 0.9947 | 0.74 |
0.837 | 4.0 | 452 | 0.8767 | 0.81 |
0.5078 | 5.0 | 565 | 0.7830 | 0.86 |
0.3832 | 6.0 | 678 | 0.7838 | 0.84 |
0.3902 | 7.0 | 791 | 0.8064 | 0.83 |
0.3322 | 8.0 | 904 | 0.7964 | 0.82 |
0.3455 | 9.0 | 1017 | 0.7507 | 0.87 |
0.2924 | 10.0 | 1130 | 0.8073 | 0.86 |
0.2925 | 11.0 | 1243 | 0.7269 | 0.86 |
0.2853 | 12.0 | 1356 | 0.7186 | 0.86 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3