w2v-bert-2.0-vietnamese-CV16.0
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the common_voice_16_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.7915
- Wer: 0.3550
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 5.3839 | 2.26 | 200 | 4.0267 | 0.9994 |
| 3.3169 | 4.52 | 400 | 3.4490 | 1.0474 |
| 2.1073 | 6.78 | 600 | 1.0066 | 0.5568 |
| 0.201 | 9.04 | 800 | 0.7915 | 0.3550 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1
- Downloads last month
- 72
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for trick4kid/w2v-bert-2.0-vietnamese-CV16.0
Base model
facebook/w2v-bert-2.0