learning_rate: 0.0003
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
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: 500
num_epochs: 30
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Cer
3.82	3.6364	400	0.8073	0.2416
1.0023	7.2727	800	0.6507	0.2112
0.8591	10.9091	1200	0.5511	0.2018
0.8033	14.5455	1600	0.5025	0.1890
0.7482	18.1818	2000	0.4685	0.1747
0.708	21.8182	2400	0.4347	0.1617
0.6821	25.4545	2800	0.4259	0.1585
0.6645	29.0909	3200	0.4073	0.1540

colorlessideas/mms-1bl1107-mocho-120

mms-1bl1107-mocho-120