superb_si_42
This model is a fine-tuned version of facebook/hubert-xlarge-ll60k on the superb dataset. It achieves the following results on the evaluation set:
- Loss: 2.7941
- Accuracy: 0.3591
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: 32
- eval_batch_size: 1
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 6.9326 | 1.0 | 4324 | 7.0085 | 0.0012 |
| 5.721 | 2.0 | 8648 | 6.2112 | 0.0055 |
| 5.0616 | 3.0 | 12972 | 5.2761 | 0.0426 |
| 4.064 | 4.0 | 17296 | 4.2350 | 0.1221 |
| 3.5838 | 5.0 | 21620 | 3.8063 | 0.1718 |
| 3.2102 | 6.0 | 25944 | 3.4211 | 0.2402 |
| 2.8328 | 7.0 | 30268 | 3.2387 | 0.2706 |
| 2.6067 | 8.0 | 34592 | 2.9958 | 0.3195 |
| 2.4651 | 9.0 | 38916 | 2.8434 | 0.3546 |
| 2.3548 | 10.0 | 43240 | 2.7941 | 0.3591 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 3
Model tree for jialicheng/superb-si_hubert-xlarge
Base model
facebook/hubert-xlarge-ll60k