mms-1b-all-bigc-combined-10hrs-52

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

• Loss: 1.3599
• Wer: 0.6873

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: 100
• num_epochs: 30.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
5.986	0.3226	100	2.1309	0.8729
0.8284	0.6452	200	1.5173	0.7702
0.7447	0.9677	300	1.4704	0.7419
0.7179	1.2903	400	1.5090	0.7080
0.7307	1.6129	500	1.3190	0.7037
0.6671	1.9355	600	1.4448	0.6962
0.6696	2.2581	700	1.4347	0.6989
0.6552	2.5806	800	1.3599	0.6871

Framework versions

• Transformers 4.53.0.dev0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.0