mms-1b-all-swagen-combined-m50f50-dnn-52-0.06

This model is a fine-tuned version of facebook/mms-1b-all on the SWAGEN - DEB dataset. It achieves the following results on the evaluation set:

• Loss: 14.5794
• Cer: 6.1232

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: 52
• 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_steps: 250
• num_epochs: 15.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
14.8440	1.9723	250	14.5793	6.1222
17.6702	3.9407	500	14.5798	6.1230
12.4657	5.9091	750	14.5799	6.1238
12.9189	7.8775	1000	14.5798	6.1231

Framework versions

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