superb_si_42
This model is a fine-tuned version of facebook/wav2vec2-large on the superb dataset. It achieves the following results on the evaluation set:
- Loss: 3.5950
- Accuracy: 0.2135
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 |
|---|---|---|---|---|
| 6.176 | 1.0 | 4324 | 6.4585 | 0.0054 |
| 5.6128 | 2.0 | 8648 | 5.8162 | 0.0127 |
| 5.2909 | 3.0 | 12972 | 5.4375 | 0.0306 |
| 4.8153 | 4.0 | 17296 | 5.0332 | 0.0553 |
| 4.4768 | 5.0 | 21620 | 4.6709 | 0.0814 |
| 4.1816 | 6.0 | 25944 | 4.2882 | 0.1191 |
| 3.8911 | 7.0 | 30268 | 4.0059 | 0.1563 |
| 3.7094 | 8.0 | 34592 | 3.7603 | 0.1910 |
| 3.5673 | 9.0 | 38916 | 3.6491 | 0.2071 |
| 3.4742 | 10.0 | 43240 | 3.5950 | 0.2135 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 15
Model tree for jialicheng/superb-si_wav2vec2-large
Base model
facebook/wav2vec2-large