superb_ks_42
This model is a fine-tuned version of facebook/wav2vec2-base on the superb dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0853
- Accuracy: 0.9846
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: 4
- 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 |
| 1.25 |
1.0 |
1597 |
0.1755 |
0.9582 |
| 0.2502 |
2.0 |
3194 |
0.1402 |
0.9713 |
| 0.2218 |
3.0 |
4791 |
0.0956 |
0.9803 |
| 0.1746 |
4.0 |
6388 |
0.0917 |
0.9797 |
| 0.17 |
5.0 |
7985 |
0.0893 |
0.9807 |
| 0.1431 |
6.0 |
9582 |
0.0933 |
0.9810 |
| 0.1238 |
7.0 |
11179 |
0.0958 |
0.9831 |
| 0.116 |
8.0 |
12776 |
0.0970 |
0.9834 |
| 0.0995 |
9.0 |
14373 |
0.0853 |
0.9846 |
| 0.0985 |
10.0 |
15970 |
0.0829 |
0.9838 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
