ssc-aln-mms-model-mix-adapt-max
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0180
- Cer: 0.2065
- Wer: 0.5208
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: 0.001
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- 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: 100
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer | Wer |
|---|---|---|---|---|---|
| 1.0191 | 0.8282 | 200 | 1.1309 | 0.2203 | 0.5700 |
| 1.0289 | 1.6542 | 400 | 1.0621 | 0.2145 | 0.5482 |
| 0.9919 | 2.4803 | 600 | 1.0266 | 0.2151 | 0.5449 |
| 0.8688 | 3.3064 | 800 | 1.0741 | 0.2102 | 0.5312 |
| 0.8974 | 4.1325 | 1000 | 1.0375 | 0.2079 | 0.5253 |
| 0.8544 | 4.9607 | 1200 | 1.0180 | 0.2065 | 0.5208 |
Framework versions
- Transformers 4.57.2
- Pytorch 2.9.1+cu128
- Datasets 3.6.0
- Tokenizers 0.22.0