metadata
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: w2v2-shallow-bart-5k-1e-3
results: []
w2v2-shallow-bart-5k-1e-3
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.2228
- Wer: 219.0067
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: 0.001
- 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 |
|---|---|---|---|---|
| 5.4271 | 0.2 | 1000 | 5.5063 | 287.8938 |
| 4.9304 | 1.04 | 2000 | 5.4201 | 198.6618 |
| 4.6411 | 1.24 | 3000 | 5.2925 | 332.8481 |
| 4.3797 | 2.09 | 4000 | 5.2913 | 155.6744 |
| 4.2848 | 2.29 | 5000 | 5.2228 | 219.0067 |
Framework versions
- Transformers 4.25.0.dev0
- Pytorch 1.11.0
- Datasets 2.6.1
- Tokenizers 0.13.2