metadata
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: testing_pretrained_niger_mali
results: []
testing_pretrained_niger_mali
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.9245
- Wer: 0.8889
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 350
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 6.427 | 35.29 | 300 | 2.9588 | 1.0 |
| 2.8653 | 70.59 | 600 | 2.7466 | 1.0 |
| 2.7675 | 105.88 | 900 | 2.7207 | 1.0 |
| 2.6674 | 141.18 | 1200 | 2.2285 | 1.0 |
| 1.7813 | 176.47 | 1500 | 1.5717 | 0.8852 |
| 1.0447 | 211.76 | 1800 | 1.7009 | 0.8778 |
| 0.8167 | 247.06 | 2100 | 1.8010 | 0.8815 |
| 0.7059 | 282.35 | 2400 | 1.8748 | 0.8815 |
| 0.6572 | 317.65 | 2700 | 1.9245 | 0.8889 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3