ssc-bew-mms-model-mix-adapt-max-lowlr

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6504
• Cer: 0.1753
• Wer: 0.5182

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: 2
• eval_batch_size: 6
• 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
0.6085	0.9828	200	0.6664	0.1788	0.5260
0.6004	1.9631	400	0.6589	0.1767	0.5229
0.5616	2.9435	600	0.6559	0.1768	0.5230
0.5716	3.9238	800	0.6498	0.1761	0.5227
0.5525	4.9042	1000	0.6504	0.1753	0.5182

Framework versions

• Transformers 4.57.2
• Pytorch 2.9.1+cu128
• Datasets 3.6.0
• Tokenizers 0.22.0