hubert-base-ft-keyword-spotting
This model is a fine-tuned version of facebook/hubert-base-ls960 on the superb dataset. It achieves the following results on the evaluation set:
- Loss: 0.0774
- Accuracy: 0.9819
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 0
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0422 | 1.0 | 399 | 0.8999 | 0.6918 |
| 0.3296 | 2.0 | 798 | 0.1505 | 0.9778 |
| 0.2088 | 3.0 | 1197 | 0.0901 | 0.9816 |
| 0.202 | 4.0 | 1596 | 0.0848 | 0.9813 |
| 0.1535 | 5.0 | 1995 | 0.0774 | 0.9819 |
Framework versions
- Transformers 4.12.0.dev0
- Pytorch 1.9.1+cu111
- Datasets 1.14.0
- Tokenizers 0.10.3
- Downloads last month
- 11