metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- common_voice
model-index:
- name: test-model-lg-data
results: []
test-model-lg-data
This model is a fine-tuned version of Monsia/test-model-lg-data on the common_voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.3151
- Wer: 0.3994
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: 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: 200
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0599 | 0.67 | 100 | 0.3357 | 0.4231 |
| 0.0749 | 1.35 | 200 | 0.4297 | 0.4959 |
| 0.1072 | 2.03 | 300 | 0.3807 | 0.4782 |
| 0.083 | 2.7 | 400 | 0.3491 | 0.4559 |
| 0.0668 | 3.38 | 500 | 0.3464 | 0.4376 |
| 0.058 | 4.05 | 600 | 0.3339 | 0.4159 |
| 0.0588 | 4.73 | 700 | 0.3151 | 0.3994 |
Framework versions
- Transformers 4.11.3
- Pytorch 1.10.0+cu113
- Datasets 1.13.3
- Tokenizers 0.10.3