w2v-bert-2.0-Swahili-CV-train-8.0
This model is a fine-tuned version of facebook/w2v-bert-2.0 on the common_voice_8_0 dataset. It achieves the following results on the evaluation set:
- Loss: inf
- Wer: 0.1762
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: 64
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- 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 |
|---|---|---|---|---|
| 1.3054 | 1.95 | 300 | inf | 0.1116 |
| 0.1079 | 3.91 | 600 | inf | 0.1036 |
| 0.0821 | 5.86 | 900 | inf | 0.0918 |
| 0.0959 | 7.82 | 1200 | inf | 0.2150 |
| 0.3709 | 9.77 | 1500 | inf | 0.1762 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.2.1+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2