learning_rate: 1e-05
train_batch_size: 8
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use 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: 300
training_steps: 3000
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
2.9347	0.0833	250	1.1584	59.4835
0.8705	0.1667	500	0.7921	46.6928
0.7898	0.25	750	0.7381	43.4892
0.6827	0.3333	1000	0.7021	38.4401
0.6178	1.0833	1250	0.6791	39.6489
0.5425	1.1667	1500	0.6660	37.5250
0.5605	1.25	1750	0.6646	36.9325
0.5056	1.3333	2000	0.6535	34.7832
0.4835	2.0833	2250	0.6488	36.5607
0.4436	2.1667	2500	0.6465	35.5098
0.4746	2.25	2750	0.6453	35.3604

wwwtwwwt/whisper-tiny-Education

whisper-tiny-Education