learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 8
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: 10
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
0.7149	1.0	42	0.6665
0.5745	2.0	84	0.6567
0.5488	3.0	126	0.6372
0.5246	4.0	168	0.6357
0.5142	5.0	210	0.6361
0.5021	6.0	252	0.6208
0.4967	7.0	294	0.6228
0.4921	8.0	336	0.6129
0.4843	9.0	378	0.6071
0.48	10.0	420	0.6053
0.4724	11.0	462	0.6031
0.47	12.0	504	0.6014
0.4671	13.0	546	0.6001
0.4629	14.0	588	0.5935
0.4551	15.0	630	0.5925
0.4568	16.0	672	0.5923
0.4549	17.0	714	0.5885
0.4589	18.0	756	0.5884
0.4489	19.0	798	0.5892
0.4522	19.5350	820	0.5864

kavinda123321/speecht5_finetuned_mahinda8_sinhala

speecht5_finetuned_mahinda8_sinhala