mms-1b-all-swagen-combined-m50f50-dnn-42-0.04

This model is a fine-tuned version of csikasote/mms-1b-all-swagen-combined-m50f50-dnn-42-0.04 on the SWAGEN - FUS dataset. It achieves the following results on the evaluation set:

• Loss: 0.3241
• Cer: 0.0789

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: 8
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• 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_ratio: 0.05
• lr_scheduler_warmup_steps: 100
• num_epochs: 30.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.4391	0.7905	100	0.4093	0.0927
0.6661	1.5771	200	0.3816	0.0872
0.0181	2.3636	300	0.3057	0.0738
0.2996	3.1502	400	0.4619	0.1165
0.2144	3.9407	500	0.3101	0.0744
0.0264	4.7273	600	0.3241	0.0790

Framework versions

• Transformers 4.53.0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.4