learning_rate: 5e-06
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant_with_warmup
lr_scheduler_warmup_steps: 50
training_steps: 500

Training Loss	Epoch	Step	Validation Loss	Global Wer
1.1045	0.0110	50	0.8921	82.8339
0.6134	0.0220	100	0.6695	68.7151
0.6054	0.0330	150	0.5966	62.8238
0.4919	0.0440	200	0.5574	64.8553
0.4787	0.0550	250	0.5505	60.5891
0.4944	0.0660	300	0.5125	55.8659
0.5035	0.0770	350	0.4955	57.0848
0.4886	0.0880	400	0.4735	52.5648
0.4687	0.0990	450	0.4566	57.0340
0.4018	0.1100	500	0.4375	58.0498

dianavdavidson/wh_small_iv_indic_voices_51708_trial

wh_small_iv_indic_voices_51708_trial