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
2.0581	0.0110	50	1.5440	69.2737
0.9198	0.0220	100	0.9594	102.1331
0.4407	0.0330	150	0.4433	66.5820
0.3859	0.0440	200	0.4062	53.7836
0.3548	0.0550	250	0.4094	50.5841
0.3782	0.0660	300	0.3786	52.3108
0.3932	0.0770	350	0.3711	64.1442
0.4017	0.0880	400	0.3572	50.4317
0.3387	0.0990	450	0.3653	75.4190
0.2846	0.1100	500	0.3423	63.3824

dianavdavidson/wh_med_iv_indic_voices_51708_trial

wh_med_iv_indic_voices_51708_trial