learning_rate: 5e-05
train_batch_size: 8
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: linear
num_epochs: 20
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
0.4185	1.0	79	0.2389	0.1684	0.8656	0.2820
0.2479	2.0	158	0.1217	0.3488	0.9175	0.5055
0.1190	3.0	237	0.0912	0.6511	0.9286	0.7655
0.1009	4.0	316	0.0789	0.6148	0.9333	0.7413
0.0659	5.0	395	0.0702	0.7041	0.9592	0.8121
0.0715	6.0	474	0.0832	0.7161	0.9490	0.8163
0.0552	7.0	553	0.0812	0.8347	0.9546	0.8906
0.0322	8.0	632	0.0816	0.8274	0.9462	0.8828
0.0224	9.0	711	0.0889	0.8292	0.9314	0.8773
0.0113	10.0	790	0.0981	0.8518	0.9110	0.8804
0.0217	11.0	869	0.0888	0.8539	0.9425	0.8960
0.0120	12.0	948	0.0927	0.8739	0.9444	0.9078
0.0117	13.0	1027	0.0808	0.8717	0.9509	0.9096
0.0051	14.0	1106	0.0969	0.8859	0.9500	0.9168
0.0062	15.0	1185	0.1082	0.8910	0.9472	0.9182
0.0083	16.0	1264	0.1011	0.8914	0.9509	0.9202
0.0071	17.0	1343	0.1053	0.8807	0.9509	0.9144
0.0145	18.0	1422	0.1068	0.8821	0.9500	0.9148
0.0035	19.0	1501	0.1039	0.8924	0.9527	0.9216
0.0026	20.0	1580	0.1060	0.8900	0.9527	0.9203

varin/lilt-infoxlm-sroie-kie

lilt-infoxlm-sroie-kie