mert-base-finetuned-gtzan
This model is a fine-tuned version of yangwang825/mert-base on the GTZAN dataset. It achieves the following results on the evaluation set:
- Loss: 0.6800
- 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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- 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 |
---|---|---|---|---|
1.0961 | 1.0 | 112 | 1.2710 | 0.59 |
0.9162 | 2.0 | 224 | 1.0297 | 0.64 |
0.721 | 3.0 | 336 | 1.1227 | 0.56 |
0.5045 | 4.0 | 448 | 0.5215 | 0.83 |
0.3727 | 5.0 | 560 | 0.5263 | 0.86 |
0.1159 | 6.0 | 672 | 0.8055 | 0.84 |
0.0276 | 7.0 | 784 | 0.5396 | 0.87 |
0.1 | 8.0 | 896 | 0.6800 | 0.88 |
0.2564 | 9.0 | 1008 | 0.5907 | 0.87 |
0.1327 | 10.0 | 1120 | 0.5915 | 0.88 |
Framework versions
- Transformers 4.25.1
- Pytorch 2.5.0+cu121
- Datasets 2.7.1
- Tokenizers 0.13.3
- Downloads last month
- 18