ssc-ttj-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: 0.1838
• Cer: 0.0697
• Wer: 0.3887

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
0.2793	0.5579	200	0.2131	0.0742	0.4136
0.2597	1.1144	400	0.2044	0.0752	0.4150
0.2649	1.6722	600	0.1958	0.0722	0.4004
0.259	2.2287	800	0.1909	0.0712	0.3978
0.2284	2.7866	1000	0.1912	0.0707	0.3946
0.2333	3.3431	1200	0.1899	0.0712	0.3981
0.2096	3.9010	1400	0.1867	0.0700	0.3902
0.2226	4.4575	1600	0.1838	0.0697	0.3887

Framework versions

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