ssc-sco-mms-model-mix-adapt-max3-devtrain

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

• Loss: 0.2287
• Cer: 0.0827
• Wer: 0.2665

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: 0.0005
• train_batch_size: 1
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 2
• 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: 100
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer	Wer
0.3174	0.9615	200	0.2341	0.0842	0.2730
0.2948	1.9231	400	0.2338	0.0846	0.2753
0.2828	2.8846	600	0.2309	0.0841	0.2729
0.2631	3.8462	800	0.2289	0.0827	0.2681
0.2362	4.8077	1000	0.2287	0.0827	0.2665

Framework versions

• Transformers 4.52.1
• Pytorch 2.9.1+cu128
• Datasets 3.6.0
• Tokenizers 0.21.4