WAXALamh_asr-w2v-bert-2.0
This model is a fine-tuned version of facebook/w2v-bert-2.0 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.8336
- Cer: 0.3028
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: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 10.4852 | 1.7241 | 300 | 2.9051 | 0.3477 |
| 1.5670 | 3.4483 | 600 | 3.2439 | 0.3243 |
| 1.1122 | 5.1724 | 900 | 3.1928 | 0.3141 |
| 0.7839 | 6.8966 | 1200 | 3.4828 | 0.3058 |
| 0.5211 | 8.6207 | 1500 | 3.7789 | 0.3047 |
| 0.5211 | 10.0 | 1740 | 3.8336 | 0.3028 |
Framework versions
- Transformers 5.12.1
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2