distilhubert-bass-classifier
This model is a fine-tuned version of ntu-spml/distilhubert on the bass_design_encoded dataset. It achieves the following results on the evaluation set:
- Loss: 0.0128
- Accuracy: 0.9984
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: 16
- eval_batch_size: 16
- 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
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.2106 | 1.0 | 2151 | 0.2989 | 0.9315 |
0.0039 | 2.0 | 4302 | 0.0544 | 0.9864 |
0.0015 | 3.0 | 6453 | 0.0928 | 0.9825 |
0.1035 | 4.0 | 8604 | 0.1823 | 0.9689 |
0.0003 | 5.0 | 10755 | 0.0330 | 0.9958 |
0.0959 | 6.0 | 12906 | 0.0915 | 0.9885 |
0.0001 | 7.0 | 15057 | 0.0163 | 0.9979 |
0.017 | 8.0 | 17208 | 0.0205 | 0.9971 |
0.0 | 9.0 | 19359 | 0.0103 | 0.9984 |
0.0 | 10.0 | 21510 | 0.0128 | 0.9984 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 68