metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: xls-r_cv_ur
results: []
xls-r_cv_ur
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: 0.5811
- Wer: 1.0
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: 8
- 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: 100
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 8.2414 | 1.14 | 100 | 2.2232 | 1.0 |
| 1.9235 | 2.27 | 200 | 1.8280 | 1.0 |
| 1.8174 | 3.41 | 300 | 1.8247 | 1.0 |
| 1.8305 | 4.55 | 400 | 1.7873 | 1.0 |
| 1.7604 | 5.68 | 500 | 1.7839 | 1.0 |
| 1.7727 | 6.82 | 600 | 1.7499 | 1.0 |
| 1.7474 | 7.95 | 700 | 1.7209 | 1.0 |
| 1.7105 | 9.09 | 800 | 1.6658 | 1.0 |
| 1.6932 | 10.23 | 900 | 1.5940 | 1.0 |
| 1.6064 | 11.36 | 1000 | 1.4630 | 1.0 |
| 1.4748 | 12.5 | 1100 | 1.1940 | 1.0 |
| 1.2831 | 13.64 | 1200 | 0.9712 | 1.0 |
| 1.1637 | 14.77 | 1300 | 0.8227 | 1.0 |
| 1.0745 | 15.91 | 1400 | 0.7349 | 1.0 |
| 0.9797 | 17.05 | 1500 | 0.6512 | 1.0 |
| 0.9098 | 18.18 | 1600 | 0.6124 | 1.0 |
| 0.8953 | 19.32 | 1700 | 0.5811 | 1.0 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.11.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1