<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
w2v2-dbart-5k-1e-4
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3370
- Wer: 15.0932
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 250
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.0771 | 0.2 | 1000 | 1.8878 | 64.0932 |
| 0.7272 | 0.4 | 2000 | 0.7003 | 23.8557 |
| 0.5948 | 0.6 | 3000 | 0.4765 | 14.4223 |
| 0.4597 | 0.8 | 4000 | 0.3761 | 14.1429 |
| 0.3704 | 1.0 | 5000 | 0.3370 | 15.0932 |
Framework versions
- Transformers 4.25.0.dev0
- Pytorch 1.11.0
- Datasets 2.6.1
- Tokenizers 0.13.2
