metadata
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: Model_G_S_P_Wav2Vec2
results: []
Model_G_S_P_Wav2Vec2
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.4056
- Wer: 0.5800
- Cer: 0.2434
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: 500
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 1.0772 | 4.17 | 400 | 1.8505 | 0.6597 | 0.2778 |
| 0.5563 | 8.33 | 800 | 1.9481 | 0.6481 | 0.2663 |
| 0.4008 | 12.5 | 1200 | 2.2166 | 0.6589 | 0.2767 |
| 0.285 | 16.67 | 1600 | 2.0445 | 0.6128 | 0.2535 |
| 0.2171 | 20.83 | 2000 | 2.2138 | 0.5943 | 0.2509 |
| 0.1611 | 25.0 | 2400 | 2.2532 | 0.5813 | 0.2455 |
| 0.1098 | 29.17 | 2800 | 2.4056 | 0.5800 | 0.2434 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 1.18.3
- Tokenizers 0.13.3