SALAMA_NEWMEDTT
This model is a fine-tuned version of Mubarak127/waxal-whisper-large-v3-lin_asr_new-app on an unknown dataset. It achieves the following results on the evaluation set:
- eval_loss: 0.6054
- eval_wer: 54.4839
- eval_runtime: 1570.5168
- eval_samples_per_second: 0.555
- eval_steps_per_second: 0.278
- epoch: 1.1606
- step: 1200
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: 1e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH 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: 500
- num_epochs: 2
- mixed_precision_training: Native AMP
Framework versions
- Transformers 5.14.1
- Pytorch 2.5.1+cu121
- Datasets 5.0.1
- Tokenizers 0.22.2
- Downloads last month
- 5