learning_rate: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
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: 400
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
No log	0.6289	200	0.7522	1.0	1.0
2.5786	1.2579	400	0.5931	1.0	1.0
2.5786	1.8868	600	0.5481	1.0	1.0
1.064	2.5157	800	0.5167	1.0	1.0
1.064	3.1447	1000	0.5379	1.0	1.0
0.6571	3.7736	1200	0.5092	1.0	1.0
0.6571	4.4025	1400	0.5170	1.0	1.0
0.362	5.0314	1600	0.5247	1.0	1.0
0.362	5.6604	1800	0.5866	1.0	1.0
0.1506	6.2893	2000	0.6023	1.0	1.0
0.1506	6.9182	2200	0.6527	1.0	1.0
0.0618	7.5472	2400	0.6691	1.0	1.0
0.0618	8.1761	2600	0.7418	1.0	1.0
0.0237	8.8050	2800	0.7754	1.0	1.0
0.0237	9.4340	3000	0.7970	1.0	1.0

Muzaffar786/whisper-kashmiri-full-data-ur

whisper-kashmiri-full-data-ur