learning_rate: 0.0003
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 64
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: 400
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Wer
0.0882	1.0	56	1.1967	0.0064	0.5404
0.0503	2.0	112	1.3110	0.0064	0.6217
0.0396	3.0	168	1.4180	0.0064	0.6232
0.0562	4.0	224	1.6773	0.0064	0.7554
0.0527	5.0	280	2.0603	0.0064	0.9328
0.0464	6.0	336	2.5807	0.0064	0.9055
0.0498	7.0	392	3.0902	0.0064	1.1502
0.0272	8.0	448	3.6158	0.0064	1.1303
0.0021	9.0	504	3.3782	0.0064	0.9853
0.0003	9.8288	550	3.4341	0.0064	0.9964

mdsingh2024/ap-i5JeeZWT4XT69p4xD3MpUA

ap-i5JeeZWT4XT69p4xD3MpUA