mms-1b-ckt-pooled

This model is a fine-tuned version of facebook/mms-1b-all.

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: 16
• eval_batch_size: 8
• seed: 42
• 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: 25
• num_epochs: 12
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
4.2676	0.2941	50	1.3063	0.9667	0.3346
0.7992	0.5882	100	1.0971	0.9159	0.2819
0.7067	0.8824	150	0.9256	0.8730	0.2374
0.6357	1.1765	200	0.9270	0.9032	0.2396
0.6663	1.4706	250	0.8971	0.8794	0.2409
0.6652	1.7647	300	0.8835	0.8730	0.2374

Framework versions

• Transformers 4.57.1
• Pytorch 2.11.0+cu130
• Datasets 4.8.5
• Tokenizers 0.22.2