distilhubert-bass-classifier5
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: 0.0292
- Accuracy: 0.9982
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 |
---|---|---|---|---|
0.4595 | 1.0 | 1914 | 0.7017 | 0.9218 |
0.8718 | 2.0 | 3828 | 0.4075 | 0.9733 |
0.0 | 3.0 | 5742 | 0.2594 | 0.9841 |
0.0 | 4.0 | 7656 | 0.1175 | 0.9918 |
0.0 | 5.0 | 9570 | 0.0862 | 0.9965 |
0.0 | 6.0 | 11484 | 0.0947 | 0.9956 |
0.6718 | 7.0 | 13398 | 0.3438 | 0.9877 |
0.0021 | 8.0 | 15312 | 0.0936 | 0.9953 |
0.0 | 9.0 | 17226 | 0.0909 | 0.9956 |
0.0 | 10.0 | 19140 | 0.0292 | 0.9982 |
Framework versions
- Transformers 4.39.2
- Pytorch 2.2.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 65