learning_rate: 0.001
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
lr_scheduler_warmup_steps: 50
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
1.165	1.0	16	0.6121	38.8298
0.3474	2.0	32	0.5480	34.5745
0.1575	3.0	48	0.5258	33.5106
0.0726	4.0	64	0.5692	34.0426
0.0457	5.0	80	0.6185	29.7872
0.0272	6.0	96	0.6163	30.8511
0.0287	7.0	112	0.5768	28.1915
0.0292	8.0	128	0.5774	32.4468
0.0251	9.0	144	0.5645	32.9787
0.0408	10.0	160	0.6638	31.3830

GrantRC/whisper-small-en-finetuned-fire

Whisper Small Fireground