learning_rate: 1e-05
train_batch_size: 12
eval_batch_size: 12
seed: 42
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: cosine
lr_scheduler_warmup_ratio: 0.05
training_steps: 4000
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2115	1.0	199	0.1916	0.0	0.0	0.0	0.0
0.1031	2.0	398	0.0947	0.8443	0.4366	0.5756	0.4440
0.0544	3.0	597	0.0634	0.9580	0.5541	0.7021	0.5660
0.0365	4.0	796	0.0489	0.9664	0.6995	0.8116	0.7058
0.0316	5.0	995	0.0637	0.9686	0.6259	0.7605	0.6362
0.0291	6.0	1194	0.0651	0.9765	0.6016	0.7445	0.6145
0.0254	7.0	1393	0.0710	0.9778	0.6132	0.7537	0.6262
0.0219	8.0	1592	0.0619	0.9766	0.6288	0.7651	0.6418

Giannis17/layoutlmv3-finetuned-invoice_ConControl_Easy2_cosine_w_r005_w_d01

layoutlmv3-finetuned-invoice_ConControl_Easy2_cosine_w_r005_w_d01