Unose-Huen-Small-Gen2
This model is a fine-tuned version of igunosalabs/Unose-Huen-Small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5754
- Wer: 56.5212
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: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 150
- training_steps: 1200
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 138.7028 | 2.3284 | 200 | 1.5035 | 56.9308 |
| 135.5995 | 4.6569 | 400 | 1.5668 | 57.0459 |
| 113.8569 | 6.9853 | 600 | 1.5537 | 57.6219 |
| 104.0199 | 9.3050 | 800 | 1.5739 | 56.4572 |
| 108.4945 | 11.6334 | 1000 | 1.5776 | 56.5852 |
| 107.0664 | 13.9619 | 1200 | 1.5754 | 56.5212 |
Framework versions
- Transformers 5.20.2
- Pytorch 2.11.0+cu130
- Datasets 4.8.5
- Tokenizers 0.23.3
- Downloads last month
- 63