learning_rate: 0.0001
train_batch_size: 4
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
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: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
0.9639	0.9639	10	0.7799
0.8011	1.9639	20	0.6695
0.7549	2.9639	30	0.6462
0.7131	3.9639	40	0.6086
0.6548	4.9639	50	0.5548
0.5983	5.9639	60	0.5237
0.5618	6.9639	70	0.4978
0.5547	7.9639	80	0.4905
0.5479	8.9639	90	0.4727
0.5284	9.9639	100	0.4907
0.5189	10.9639	110	0.4742
0.5166	11.9639	120	0.4603
0.5056	12.9639	130	0.4541
0.5127	13.9639	140	0.4897
0.4959	14.9639	150	0.4633
0.4939	15.9639	160	0.4496
0.4649	16.9639	170	0.4403
0.4672	17.9639	180	0.4327
0.461	18.9639	190	0.4349
0.4558	19.9639	200	0.4271

kavinda123321/speecht5_mahinda_work_aug

speecht5_mahinda_work_aug