test3
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 140.4698
- Cer: 1.7882
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- 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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 101.7886 | 1.0 | 3 | 141.5133 | 1.8588 |
| 76.3259 | 2.0 | 6 | 141.5047 | 1.7882 |
| 82.7714 | 3.0 | 9 | 141.4292 | 1.8118 |
| 107.6911 | 4.0 | 12 | 141.2027 | 1.6706 |
| 81.6931 | 5.0 | 15 | 140.4698 | 1.7882 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.10.0+cu113
- Datasets 2.20.0
- Tokenizers 0.15.2
- Downloads last month
- 2